A Quality Improvement Project to Improve the Pathway and Outcomes for Patients with Necrotising Fasciitis

Necrotising fasciitis (NF) is a rare, life-threatening, and rapidly progressing disease of the soft tissue (fascia and subcutaneous tissue) that requires early surgical debridement, broad spectrum antibiotics, and systemic support. Mortality rates of between 6% to 76% have been reported^1-4 and an estimated 500 cases of NF occur each year in the United Kingdom.⁵ Prompt diagnosis and treatment of NF is associated with lower mortality rates⁶, and a multidisciplinary approach among surgeons in multiple specialties, intensivists, and microbiologists is vital. Managing patients with NF is a challenging, onerous, and time-intensive process. Multiple serial debridements, dressing changes, and stabilization of the patient are required prior to reconstruction of the debrided areas. The authors’ institution is a regional referral centre within the National Health Services (NHS) framework for patients with NF. Referrals are generally divided into regional referrals (patients transferred from peripheral hospitals to the institution for reconstruction postsurgical debridement) or internal referrals (patients admitted to the unit through the emergency department).

The aim of the study was to retrospectively analyze the case workload among the cohort in order to determine the burden of disease in the hospital and streamline services  to secure adequate resources to manage these complex patients more efficiently.

A retrospective single-centre review of 41 consecutive adult patients referred to the department with a confirmed diagnosis of NF was conducted. Patients were identified through clinical coding and cross-checked against the departmental inpatient handover list. Data were collected on patients referred to the plastic surgery department at Southmead Hospital between January 2017 and September 2018 over a period of 21 months. Referrals were categorized into either regional or internal depending on whether the initial debridement had taken place at a peripheral hospital or conducted internally on patients admitted through the emergency department. Diagnosis of NF was confirmed through either operative or histopathological findings. Patients were also further subclassified into body map regions depending on the location of infection.

Demographic data including age, gender, comorbidities, infectious organisms, and mortality were collected. Operative findings and procedures that were conducted were obtained from the electronic patient records. Data on hospital length of stay (LOS), intensive care LOS, number of operative procedures, and length of time of each operative procedure were obtained by the business intelligence team from computerized records using ICD-10 codes.

Demographics

A total of 41 patients with a confirmed diagnosis of NF (24 internal referrals and 17 regional referrals) were identified. There were 23 male and 18 female patients with a median age of 63 years (range, 28 to 85 years). The most common comorbidity was cardiovascular disease (56%). Other comorbidities included obesity (10%), diabetes mellitus (34%), renal disease (10%), and malignancy (17%). Fourteen patients had Type 2 diabetes (9 internal referrals and 5 external referrals). The mortality rates among internal and regional referrals were 38% (9 of 24) and 11% (2 of 17), respectively. The overall mortality rate within the hospital was 27% (11 of 41). A total of 177 operations were conducted (80 procedures by plastic surgery, 36 by general surgery and 5 by vascular surgery).

Regional Referrals

Among the 17 regional referrals, comorbidities included cardiovascular disease (47%), malignancy (18%), renal disease (6%) and diabetes mellitus (29%). Infectious organism data could not be obtained for this group. The median LOS was 14 days (range, 1 to 48 days). Only 2 of the patients were admitted to the intensive care unit (ICU) for 7 days each. A total of 56 operations were performed in this cohort of patients. The median number of operations per patient was 2 (range 1 to 8) and the median total theatre time per patient was 306 minutes (range, 50 to 1164 minutes). The median time per operation was 125 minutes (range, 50 to 328 minutes).

Categorically by body region, there were 4 upper limb, 6 lower limb, 2 chest/back, 1 abdominal, and 4 perianal/perineum cases. Sixteen (94%) of 17 patients underwent reconstruction with split thickness skin grafts, and 1 patient had a free-flap reconstruction. None of the regional referral patients had a defunctioning procedure.

Internal Referrals

Among the 24 internal referrals, comorbidities included cardiovascular disease (63%), malignancy (17%), renal disease (13%), and diabetes mellitus (37%). The most common infectious organism isolated was group A streptococcus in 33% of cases followed by Staphylococcus aureus in 21%. Polymicrobial infections accounted for 33% of patients, and 63% (5 of 8) of patients with diabetes had polymicrobial infections. The median LOS was 25 days (range, 11 to 94 days), and median ICU LOS was 8 days (range, 0 to 64 days). A total of 121 operations were performed in this group. The median number of operations per patient was 5 (range 1 to 14) and the median total theatre time per patient was 662 minutes (range 171 to 3007 minutes). The median time per operation was 145 minutes (range 50 to 605 minutes).

Categorically by body region, there were 5 upper limb, 7 lower limb, 3 chest/back, 3 abdominal, and 6 perianal/perineum cases. Out of the 11 (46%) of 24 patients who underwent reconstruction with split thickness skin grafts, there was 1 mortality. The median time between initial debridement to grafting was 13 days (n = 11; range, 2 to 38 days). In addition, 1 local flap, 2 pedicled flaps and 1 free flap were used to reconstruct defects. Six patients underwent defunctioning stoma procedures.

Categorical analysis according to the 4 body regions of internal referrals is summarized by bar charts, including median LOS, median ICU LOS, median number of operations, and median total theatre time per patient. (Figure 1)

NF patients who are managed by the internal department are admitted in 2 ways. Regional referrals are patients who have been diagnosed with NF and have already been debrided at a peripheral hospital in the region. Referrals are triaged by the on-call plastic surgery consultant and are based on clinical details and photos supplied by the referring clinician. Advice is given on whether further debridement is required and the types of dressing to use, including vacuum-assisted closure (VAC) dressings. Patients are transferred to the department for reconstructive surgery when clinical photos confirm adequate debridement, and the patient is medically stable and safe to transfer. Transfer is delayed until the patient no longer requires ICU support and has had a defunctioning procedure, if indicated, for perineal defects.

Internal referrals are patients with NF or suspected NF that are referred to the department either through the emergency department or another speciality within the hospital. The plastic surgery team is consulted for a clinical review and, depending on the clinical and biochemical findings, NF patients are identified and management is initiated. Initial debridement is performed by body region and is assigned to specific surgical specialties. Upper limbs are assigned to plastic surgery/orthopaedic surgery; lower limbs are assigned to orthopaedic surgery; head and neck  is assigned to plastic surgery/ear; nose, and throat (ENT)/oral and maxillofacial surgery (OMFS); and chest/trunk/perianal/perineum are assigned to general surgery/urology. Advice and assistance are provided by the team regarding dressings and the extent of debridement required (intraoperatively). Patients are generally transferred to the department or co-managed when the patient and their wounds are ready for reconstruction. Cases are discussed daily, and patients are managed accordingly in conjunction with the ICU, microbiology, and parent teams. Reconstruction is generally delayed until the patient has been stabilized on intensive care, vasopressors have been discontinued, there are no further signs of ongoing infection or necrosis, there is recovery from the catabolic phase of the disease with an increasing trend in the serum albumin (with nutritional support, as required), transfusions for anaemia (if required) have been given and fashioning of a defunctioning stoma by the general surgeons in perianal/perineal cases. Patients with disease affecting the perianal/perineum region had an increased median number of operations and median total theatre time most likely corresponding to the need for a defunctioning stoma prior to reconstruction and the complexity of operating in this region in terms of positioning and application of dressings. Interestingly, median LOS was substantially longer in abdominal wounds. This was likely due to complex wounds involving laparotomies that were performed in 2 of the 3 patients.

The results of this study have helped define the workload involved when managing patients with NF. Patients with NF have extremely complex requirements. Having data on the number of theatre visits and theatre time required per patient have helped to logistically manage the case load. Management planning was improved by negotiating extra theatre lists from hospital managers and allocating enough surgeons in the theatre to make the process more efficient. Dressing changes and skin grafting of large areas are time-intensive processes requiring a large number of personnel, and a team-based approach is vital in ensuring that the procedures are carried out as efficiently as possible. Patients with NF who have had operations performed by the team are generally allocated to the plastic surgery planned daily trauma lists, and this has a knock-on effect by delaying other trauma workloads including hand and lower limb cases. Being able to forecast the need for increased theatre capacity has helped to plan and allocate valuable resources in order to manage NF patients alongside the acute trauma workload. In some cases, this may involve taking on planned elective lists in order to accommodate the trauma workload alongside the NF patients.

Apart from the strain on theatre capacity and staffing levels required, the management of patients with NF carries a huge financial burden to the NHS. It was beyond the scope of this study to calculate the exact costs per patient, but theatre costs per hour can be estimated at around $1600 (£1200 GBP) per hour.⁷ Extrapolating from this, median theatre cost per patient for internal referrals was estimated to be $18000 (£13 240 GBP) and $8500 (£6120 GBP) for regional referrals. Other costs, however, should be considered, including ICU stay per day, surgical ward bed, dressings, and rehabilitation. Results from a study conducted in Australia between 2000 and 2004 showed that the mean cost per patient for in-patient treatment of NF alone was $48500 ($64517 AUD ).⁸ When comparing regional referrals against internal referrals, the median LOS was 44% lower, the median number of operations per patient was 60% lower, and median total theatre time per patient was 53% lower, indicating that the initial debridement and critical care phase accounts for approximately 50% of the workload with the reconstructive phase accounting for the remaining half.

The overall mortality rate of 27% in this study is comparable to other reports.² The variation between mortality among internal referrals (38%) and regional referrals (11%) was due to patients being medically stable enough to be transferred to the facility for reconstruction, having survived the initial debridement phase. No data was available on any pretransfer mortalities in other institutions for comparison. Among the internal referral mortalities, 44% of the patients had diabetes and all 9 patients had multiple comorbidities and an average age of 74 years. In conjunction, these factors were also very likely to contribute to the higher mortality rate among the internal referral cohort. Inclusion or exclusion criteria were not specified, and all patients with a diagnosis of NF were included in the review. The mortality rate in 3302 NF patients between 1980 and 2008 was 23.5% and this figure is trending downward; however, it remains above 20%.⁹ NF remains a complex and challenging condition to treat, and adequate surgical debridement, broad spectrum antibiotics, and input from intensive care specialists are all vital.^10,11 Early aggressive surgical debridement has been shown to reduce mortality in NF patients when conducted within the first 24 hours.^6,12

Data on theatre time was extracted from the theatre management system and did not include theatre turnaround time and anaesthetic induction time, which can be quite lengthy and complex for these patients and result in a theatre session time of 3.5 hours.¹³ The type of procedure was not taken into consideration when determining the median theatre time and all data were pooled together. For example, dressing changes and skin grafting procedures were not categorized or analyzed independently. Total body surface area of the defect was not considered as an independent variable in the reconstructive process. Internal referral time between presentation and initial debridement was not analyzed. Due to the NHS framework, accurate real treatment costs could not be calculated and could only be extrapolated. Being a single-centre retrospective case series, the data were used as part of a quality improvement project locally and may not be generally applied.

Early diagnosis and initiation of immediate treatment in patients with NF may provide the best opportunity for a positive outcome.¹⁴ NF patients require complex medical and surgical care that can be heavily resource dependent. New technologies, such as autologous full-thickness skin grafts, may further increase efficiency by reducing the number of operative procedures and overall costs, and may provide better outcomes for these patients compared to traditional methods.¹⁵ By characterising and defining the workload involved in treating NF patients, it may be possible to improve advanced planning and allocate resources accordingly in order to ensure that the centre runs as smoothly as possible without interruption to the acute trauma service.

Affiliations: North Bristol NHS Trust, Southmead Hospital, Southmead Road, Bristol, BS10 5NB

Correspondence: M Thakkar, North Bristol NHS Trust, Southmead Hospital, Southmead Road, Bristol, BS10 5NB; m.thakkar@nhs.net

Disclosure: The authors disclose no financial or other conflicts of interest.