PLASTICS ANZJSurg.com
Are we getting necrotizing soft tissue infections right? A 10-year review David Proud,* Frank Bruscino Raiola,* Dane Holden,* Eldho Paul,† Robert Capstick* and Amy Khoo‡ *Plastic, Hand and Maxillofacial Surgery Unit, Alfred Hospital, Melbourne, Victoria, Australia †Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia and ‡Department of Surgery, Monash Health, Melbourne, Victoria, Australia
Key words Fournier’s gangrene, necrotizing fasciitis Correspondence Dr David Proud, Plastic, Hand and Maxillofacial Surgery Unit, Alfred Hospital, Commercial Road, Prahran, Vic. 3181, Australia. Email: [email protected] D. Proud MBBS; F. Bruscino Raiola MBBS, FRACS; D. Holden MBBS; E. Paul MSc; R. Capstick MBBS; A. Khoo MBBS. Accepted for publication 6 September 2013. doi: 10.1111/ans.12412
Abstract Background: The Alfred Hospital is a referral centre for necrotizing soft tissue infections (NSTIs) in the state of Victoria and receives around 20 such patients each year. We sought to compare our practice and outcomes against published data, and to examine management at referring hospitals to determine whether adjustments to current practices are required. Methods: A retrospective chart review of patients admitted to the Alfred Hospital between 1 January 2001 and 31 December 2010 with a diagnosis of necrotizing fasciitis was conducted. Demographic, etiologic, treatment and outcome data were collected and analysed. Results: Two hundred and nineteen patients were identified with a mean age of 54.76 years and a preponderance of men (63.47%). The overall mortality rate for the patient group was 15.98%. More than 80% of patients were transferred from another facility. Nearly 40% of patients did not undergo surgical debridement within 24 h of presentation to a hospital and 30.6% were not debrided prior to transfer. Patients underwent a median of three procedures at the Alfred Hospital and the majority of patients required admission to the intensive care unit (68.95%). Conclusion: NSTIs remain a surgical emergency with high rates of mortality and resource requirements. The mortality rate at our institution compares well with other published series. Many patients experienced delays before undergoing debridement and in many cases were transferred without debridement. The trend towards transferring NSTI patients to centres accustomed to treating burns and major trauma seems logical, but should not delay life-saving surgical debridement. Timing of transfer does not seem to affect mortality.
Introduction Necrotizing soft tissue infections (NSTIs) are a rare but serious surgical emergency. Mortality rates quoted in previous observational studies have ranged from 6–72% and the main predictor of mortality is time to surgical debridement and initiation of appropriate antibiotic therapy.1 There is ongoing debate as to the benefits of treating patients with NSTIs with hyperbaric oxygen therapy (HBO), but current consensus is that benefits are as an adjunct to surgery.2 The Alfred Hospital is the home of the state of Victoria’s adult burns service and a major trauma referral centre, as well as the site of the state’s only publicly funded hyperbaric oxygen chamber. Our ANZ J Surg 84 (2014) 468–472
facility is referred to many patients with NSTIs each year for specialist management, including HBO. This study was set out to identify demographic data, and quantify resource allocation and outcomes for patients admitted to the Alfred Hospital with NSTIs. Data were also collected on management prior to transfer to our facility. In our facility, patients with NSTIs affecting the extremities are admitted under the Department of Plastic and Reconstructive Surgery and those affecting the trunk are admitted under the Department of General Surgery. All patients undergo at least one debridement within 24 h of admission and thereafter as considered necessary. After debridement is complete the plastic surgery team © 2013 Royal Australasian College of Surgeons
Necrotizing soft tissue infection review
perform reconstruction as needed. All patients are referred to the hyperbaric medicine unit and undergo HBO if deemed appropriate. Antibiotic therapy is directed by consultation with the infectious diseases unit.
Methods Setting The state of Victoria, Australia, has a population of approximately 5.35 million, more than two-thirds of which reside in the capital city of Melbourne.3 The Alfred Hospital is a 390-bed university-affiliated tertiary adult-medicine referral hospital located near the city centre.
Ethics Ethics approval was obtained from the Alfred Hospital Ethics Committee.
Data collection Retrospective chart analysis of all patients with discharge International Classification of Diseases 10 coding for necrotizing fasciitis between 1 January 2001 and 31 December 2010. Two hundred and fifty-one patients were identified in this manner. Patients were included in the study if they had an operative report describing necrosis of muscle, fat or fascia, or a histopathology report confirming necrosis at any of these levels. Patients were excluded from review if they did not have a NSTI described in their medical record. This left 219 patients eligible for data collection. Data reviewed included observation charts, operation reports, pathology and radiology reports, ambulance reports and referral information from external hospitals where available. Data collected were: age, sex, co-morbidities, affected areas, diagnostic studies performed, histological and microbiological results, hospital and intensive care unit (ICU) length of stay (LOS), timing of transfer and surgical debridement after presentation, number and duration of surgical procedures, and discharge disposition. Data were also collected regarding mechanical ventilation, inotropic support, antibiotic therapy and HBO.
Data analysis All collected data were de-identified and analysed using the SAS software version 9.2 (SAS Institute, Cary, NC, USA). Continuous data are expressed as mean ± standard deviation (SD) or median (inter-quartile range (IQR)) depending on the underlying distribution of the data. Categorical data are presented as counts and proportions. Logistic regression analysis was performed to determine the risk factors for mortality with results reported as odds ratios and 95% confidence intervals. Statistical significance was set at a twosided P-value of 0.05.
Results For the period 1 January 2001 to 31 December 2010, there were 251 patients admitted to the Alfred Hospital who were coded as having had necrotizing fasciitis on discharge. Of these patients, 32 were excluded from analysis because they were not felt to have had a © 2013 Royal Australasian College of Surgeons
469
Table 1 Characteristics of patients with NSTIs treated at the Alfred Hospital 1 January 2001–31 December 2010 (n = 219) Age (years) Male Diabetes Smoker Immunosuppression
54.76 ± 17.88 139 (63.47) 75 (34.25) 51 (23.29) 29 (13.24)
Values are mean ± standard deviation or number (percentage).
Table 2 Distribution of involved areas (n = 219) (%) Head/neck Chest Abdomen Genitalia Perineum Upper limbs Lower limbs
1.37 6.39 19.82 17.35 17.81 16.89 49.32
Table 3 Diagnostic information (n = 219) Symptoms/signs Exquisite pain Rapidly spreading cellulitis Subcutaneous crepitus Radiological investigations Plain X-ray Ultrasound Computed tomography Magnetic resonance imaging
51.98 47.24 6.97 11.93 15.14 28.44 2.29
Values are percentages.
NSTI after chart review (actual diagnoses included peripheral vascular ulcers, meningococcaemia, soft tissue abscesses and cellulitis) or because they did not have their acute illness treated at the Alfred Hospital but were instead referred for reconstructive procedures. One paediatric patient was transferred to the Alfred Hospital for a single HBO treatment only and was not included in analysis. Of the remaining 219 patients, the mean age was 54.76 years (SD 17.88, range 12–90) with a preponderance of men (63.47%). Further demographic data are presented in Table 1. A majority (81.26%) of these patients were transferred to the Alfred Hospital from another facility. Of these, 38.9% were transferred within 24 h of presentation, 28.7% between 24 and 48 h, and 32.4% after 48 h.
Disease and diagnosis Table 2 shows the distribution of areas affected with the lower limb (49.32%) and abdominal wall (19.82%) being the most commonly affected areas. Prevalence of clinical signs and use of diagnostic radiological investigations are demonstrated in Table 3. Of the patients who underwent computed tomography (CT) scanning (n = 61), it was felt to aid diagnosis 55% of the time by either raising the possibility of NSTI in the report or by suspicious features (subcutaneous gas, fluid collections, oedema).
470
Proud et al.
Table 4 Microbiology Cultured organisms (n = 165)
Common infecting organisms by anatomical region (%)
Region
Number of patients
Organism (% of infections at site)
Staphylococcus aureus Streptococcus pyogenes Escherichia coli Pseudomonas aeruginosa Enterococcus spp. Clostridium perfringens
29.7 27.27 12.12 12.12 10.91 1.82
Upper limb Lower limb Abdomen Chest Genitalia Perineum
32 98 34 12 31 30
Other organisms
36.97
Head/neck
3
S. pyogenes (50) S. aureus (28.57), S. pyogenes (28.57) S. aureus (26.47), E. coli (20) S. pyogenes (41.67), S. aureus (33.33) E. coli (22.58), S. aureus (16.13) E. coli (20), S. aureus (16.67), Enterococcus (10) Group C Streptococci (100)
Microbiology At least one specific organism was cultured in 165 patients (73.06%) and infection was monomicrobial in 59.95% of these cases. The cultured organisms are demonstrated in Table 4 with Staphylococcus aureus and Streptococcus pyogenes being the most common pathogens. More than half (56.25%) of S. aureus infections were polymicrobial. In patients who developed a NSTI (n = 40) after a surgical procedure, S. aureus and Escherichia coli were the most commonly cultured organisms. Table 4 also highlights the most common pathogen isolated from each anatomical region.
Surgical and supportive treatment The majority of patients (69.4%) who were transferred to the Alfred Hospital underwent surgical debridement at the referring facility and 63.3% underwent at least one surgical debridement in the first 24 h after presentation. Patients who were surgically debrided at the Alfred Hospital experienced a median of three procedures (IQR = 2–5 procedures), and a median total operating time of 202.5 min (IQR = 117.5–360 min). The median time to the final debridement procedure was 8 days (IQR = 3–16 days). Other procedures performed included amputation (10.96%), diverting enterostomy/ colostomy (4.11%), local transposition of tissue (2.28%), free tissue transfer (2.74%) and split-thickness skin graft (47.49%). The majority of patients required admission to the ICU (68.95%) with a median LOS of 72 h (IQR = 0–238 h). Many patients arrived at the Alfred Hospital already intubated (45.71%) and 54.34% of patients required mechanical ventilation at some point during admission with a median duration of ventilation of 12 h (IQR = 0–115 h). Circulatory support with inotropic/vasopressor agents was required by 50.23% of patients. HBO was administered to 82.65% of patients with a median number of seven treatments (IQR = 3–10 treatments).
Outcomes Duration of hospital LOS at the Alfred Hospital ranged from 0.25 to 473 days (median = 21 days, IQR: 9–39). Patient disposition on discharge was either to home (30.60%), to a rehabilitation facility (23.74%), to another acute hospital (29.68%) or death (15.98%).
Mortality Analysis was performed comparing survivors (n = 184) to nonsurvivors (n = 35) in order to identify the risk factors for mortality.
A significantly higher proportion of patients in the non-survivors group required inotropic support (88.2% versus 43%, OR = 9.90, 95% CI = 3.35–29.28), arrived at the Alfred Hospital intubated (73.5% versus 40.3%, OR = 4.11, 95% CI = 1.81–9.32), required mechanical ventilation (77.1% versus 50%, OR = 3.37, 95% CI = 1.46–7.82) or ICU admission (85.7% versus 65.7%, OR = 3.12, 95% CI = 1.16–8.44). Advancing age was also found to be a significant predictor of mortality (mean age in survivors group 53 years versus 66 years in non-survivors group, OR = 1.06, 95% CI = 1.03–1.09). Median LOS was 2 days in the non-survivors group compared with 25 days in the survivors group. In patients who were transferred to the Alfred Hospital and subsequently died, a greater proportion was transferred early (
Are we getting necrotizing soft tissue infections right? A 10-year review David Proud,* Frank Bruscino Raiola,* Dane Holden,* Eldho Paul,† Robert Capstick* and Amy Khoo‡ *Plastic, Hand and Maxillofacial Surgery Unit, Alfred Hospital, Melbourne, Victoria, Australia †Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia and ‡Department of Surgery, Monash Health, Melbourne, Victoria, Australia
Key words Fournier’s gangrene, necrotizing fasciitis Correspondence Dr David Proud, Plastic, Hand and Maxillofacial Surgery Unit, Alfred Hospital, Commercial Road, Prahran, Vic. 3181, Australia. Email: [email protected] D. Proud MBBS; F. Bruscino Raiola MBBS, FRACS; D. Holden MBBS; E. Paul MSc; R. Capstick MBBS; A. Khoo MBBS. Accepted for publication 6 September 2013. doi: 10.1111/ans.12412
Abstract Background: The Alfred Hospital is a referral centre for necrotizing soft tissue infections (NSTIs) in the state of Victoria and receives around 20 such patients each year. We sought to compare our practice and outcomes against published data, and to examine management at referring hospitals to determine whether adjustments to current practices are required. Methods: A retrospective chart review of patients admitted to the Alfred Hospital between 1 January 2001 and 31 December 2010 with a diagnosis of necrotizing fasciitis was conducted. Demographic, etiologic, treatment and outcome data were collected and analysed. Results: Two hundred and nineteen patients were identified with a mean age of 54.76 years and a preponderance of men (63.47%). The overall mortality rate for the patient group was 15.98%. More than 80% of patients were transferred from another facility. Nearly 40% of patients did not undergo surgical debridement within 24 h of presentation to a hospital and 30.6% were not debrided prior to transfer. Patients underwent a median of three procedures at the Alfred Hospital and the majority of patients required admission to the intensive care unit (68.95%). Conclusion: NSTIs remain a surgical emergency with high rates of mortality and resource requirements. The mortality rate at our institution compares well with other published series. Many patients experienced delays before undergoing debridement and in many cases were transferred without debridement. The trend towards transferring NSTI patients to centres accustomed to treating burns and major trauma seems logical, but should not delay life-saving surgical debridement. Timing of transfer does not seem to affect mortality.
Introduction Necrotizing soft tissue infections (NSTIs) are a rare but serious surgical emergency. Mortality rates quoted in previous observational studies have ranged from 6–72% and the main predictor of mortality is time to surgical debridement and initiation of appropriate antibiotic therapy.1 There is ongoing debate as to the benefits of treating patients with NSTIs with hyperbaric oxygen therapy (HBO), but current consensus is that benefits are as an adjunct to surgery.2 The Alfred Hospital is the home of the state of Victoria’s adult burns service and a major trauma referral centre, as well as the site of the state’s only publicly funded hyperbaric oxygen chamber. Our ANZ J Surg 84 (2014) 468–472
facility is referred to many patients with NSTIs each year for specialist management, including HBO. This study was set out to identify demographic data, and quantify resource allocation and outcomes for patients admitted to the Alfred Hospital with NSTIs. Data were also collected on management prior to transfer to our facility. In our facility, patients with NSTIs affecting the extremities are admitted under the Department of Plastic and Reconstructive Surgery and those affecting the trunk are admitted under the Department of General Surgery. All patients undergo at least one debridement within 24 h of admission and thereafter as considered necessary. After debridement is complete the plastic surgery team © 2013 Royal Australasian College of Surgeons
Necrotizing soft tissue infection review
perform reconstruction as needed. All patients are referred to the hyperbaric medicine unit and undergo HBO if deemed appropriate. Antibiotic therapy is directed by consultation with the infectious diseases unit.
Methods Setting The state of Victoria, Australia, has a population of approximately 5.35 million, more than two-thirds of which reside in the capital city of Melbourne.3 The Alfred Hospital is a 390-bed university-affiliated tertiary adult-medicine referral hospital located near the city centre.
Ethics Ethics approval was obtained from the Alfred Hospital Ethics Committee.
Data collection Retrospective chart analysis of all patients with discharge International Classification of Diseases 10 coding for necrotizing fasciitis between 1 January 2001 and 31 December 2010. Two hundred and fifty-one patients were identified in this manner. Patients were included in the study if they had an operative report describing necrosis of muscle, fat or fascia, or a histopathology report confirming necrosis at any of these levels. Patients were excluded from review if they did not have a NSTI described in their medical record. This left 219 patients eligible for data collection. Data reviewed included observation charts, operation reports, pathology and radiology reports, ambulance reports and referral information from external hospitals where available. Data collected were: age, sex, co-morbidities, affected areas, diagnostic studies performed, histological and microbiological results, hospital and intensive care unit (ICU) length of stay (LOS), timing of transfer and surgical debridement after presentation, number and duration of surgical procedures, and discharge disposition. Data were also collected regarding mechanical ventilation, inotropic support, antibiotic therapy and HBO.
Data analysis All collected data were de-identified and analysed using the SAS software version 9.2 (SAS Institute, Cary, NC, USA). Continuous data are expressed as mean ± standard deviation (SD) or median (inter-quartile range (IQR)) depending on the underlying distribution of the data. Categorical data are presented as counts and proportions. Logistic regression analysis was performed to determine the risk factors for mortality with results reported as odds ratios and 95% confidence intervals. Statistical significance was set at a twosided P-value of 0.05.
Results For the period 1 January 2001 to 31 December 2010, there were 251 patients admitted to the Alfred Hospital who were coded as having had necrotizing fasciitis on discharge. Of these patients, 32 were excluded from analysis because they were not felt to have had a © 2013 Royal Australasian College of Surgeons
469
Table 1 Characteristics of patients with NSTIs treated at the Alfred Hospital 1 January 2001–31 December 2010 (n = 219) Age (years) Male Diabetes Smoker Immunosuppression
54.76 ± 17.88 139 (63.47) 75 (34.25) 51 (23.29) 29 (13.24)
Values are mean ± standard deviation or number (percentage).
Table 2 Distribution of involved areas (n = 219) (%) Head/neck Chest Abdomen Genitalia Perineum Upper limbs Lower limbs
1.37 6.39 19.82 17.35 17.81 16.89 49.32
Table 3 Diagnostic information (n = 219) Symptoms/signs Exquisite pain Rapidly spreading cellulitis Subcutaneous crepitus Radiological investigations Plain X-ray Ultrasound Computed tomography Magnetic resonance imaging
51.98 47.24 6.97 11.93 15.14 28.44 2.29
Values are percentages.
NSTI after chart review (actual diagnoses included peripheral vascular ulcers, meningococcaemia, soft tissue abscesses and cellulitis) or because they did not have their acute illness treated at the Alfred Hospital but were instead referred for reconstructive procedures. One paediatric patient was transferred to the Alfred Hospital for a single HBO treatment only and was not included in analysis. Of the remaining 219 patients, the mean age was 54.76 years (SD 17.88, range 12–90) with a preponderance of men (63.47%). Further demographic data are presented in Table 1. A majority (81.26%) of these patients were transferred to the Alfred Hospital from another facility. Of these, 38.9% were transferred within 24 h of presentation, 28.7% between 24 and 48 h, and 32.4% after 48 h.
Disease and diagnosis Table 2 shows the distribution of areas affected with the lower limb (49.32%) and abdominal wall (19.82%) being the most commonly affected areas. Prevalence of clinical signs and use of diagnostic radiological investigations are demonstrated in Table 3. Of the patients who underwent computed tomography (CT) scanning (n = 61), it was felt to aid diagnosis 55% of the time by either raising the possibility of NSTI in the report or by suspicious features (subcutaneous gas, fluid collections, oedema).
470
Proud et al.
Table 4 Microbiology Cultured organisms (n = 165)
Common infecting organisms by anatomical region (%)
Region
Number of patients
Organism (% of infections at site)
Staphylococcus aureus Streptococcus pyogenes Escherichia coli Pseudomonas aeruginosa Enterococcus spp. Clostridium perfringens
29.7 27.27 12.12 12.12 10.91 1.82
Upper limb Lower limb Abdomen Chest Genitalia Perineum
32 98 34 12 31 30
Other organisms
36.97
Head/neck
3
S. pyogenes (50) S. aureus (28.57), S. pyogenes (28.57) S. aureus (26.47), E. coli (20) S. pyogenes (41.67), S. aureus (33.33) E. coli (22.58), S. aureus (16.13) E. coli (20), S. aureus (16.67), Enterococcus (10) Group C Streptococci (100)
Microbiology At least one specific organism was cultured in 165 patients (73.06%) and infection was monomicrobial in 59.95% of these cases. The cultured organisms are demonstrated in Table 4 with Staphylococcus aureus and Streptococcus pyogenes being the most common pathogens. More than half (56.25%) of S. aureus infections were polymicrobial. In patients who developed a NSTI (n = 40) after a surgical procedure, S. aureus and Escherichia coli were the most commonly cultured organisms. Table 4 also highlights the most common pathogen isolated from each anatomical region.
Surgical and supportive treatment The majority of patients (69.4%) who were transferred to the Alfred Hospital underwent surgical debridement at the referring facility and 63.3% underwent at least one surgical debridement in the first 24 h after presentation. Patients who were surgically debrided at the Alfred Hospital experienced a median of three procedures (IQR = 2–5 procedures), and a median total operating time of 202.5 min (IQR = 117.5–360 min). The median time to the final debridement procedure was 8 days (IQR = 3–16 days). Other procedures performed included amputation (10.96%), diverting enterostomy/ colostomy (4.11%), local transposition of tissue (2.28%), free tissue transfer (2.74%) and split-thickness skin graft (47.49%). The majority of patients required admission to the ICU (68.95%) with a median LOS of 72 h (IQR = 0–238 h). Many patients arrived at the Alfred Hospital already intubated (45.71%) and 54.34% of patients required mechanical ventilation at some point during admission with a median duration of ventilation of 12 h (IQR = 0–115 h). Circulatory support with inotropic/vasopressor agents was required by 50.23% of patients. HBO was administered to 82.65% of patients with a median number of seven treatments (IQR = 3–10 treatments).
Outcomes Duration of hospital LOS at the Alfred Hospital ranged from 0.25 to 473 days (median = 21 days, IQR: 9–39). Patient disposition on discharge was either to home (30.60%), to a rehabilitation facility (23.74%), to another acute hospital (29.68%) or death (15.98%).
Mortality Analysis was performed comparing survivors (n = 184) to nonsurvivors (n = 35) in order to identify the risk factors for mortality.
A significantly higher proportion of patients in the non-survivors group required inotropic support (88.2% versus 43%, OR = 9.90, 95% CI = 3.35–29.28), arrived at the Alfred Hospital intubated (73.5% versus 40.3%, OR = 4.11, 95% CI = 1.81–9.32), required mechanical ventilation (77.1% versus 50%, OR = 3.37, 95% CI = 1.46–7.82) or ICU admission (85.7% versus 65.7%, OR = 3.12, 95% CI = 1.16–8.44). Advancing age was also found to be a significant predictor of mortality (mean age in survivors group 53 years versus 66 years in non-survivors group, OR = 1.06, 95% CI = 1.03–1.09). Median LOS was 2 days in the non-survivors group compared with 25 days in the survivors group. In patients who were transferred to the Alfred Hospital and subsequently died, a greater proportion was transferred early (
