burns 41 (2015) 454–461
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Merits and challenges in the development of a dedicated burn service at a regional hospital in South Africa Nikki L. Allorto *, Damian L. Clarke Department of Surgery, Pietermaritzburg Metropolitan Complex, University of Kwa-Zulu Natal, South Africa
article info
abstract
Article history:
Introduction: The Edendale Hospital Burn Service was initiated in 2011 to improve the quality of
Accepted 28 July 2014
burn care at a regional hospital. This audit reviews the merits and challenges in developing such a service and identifies areas on which to focus quality improvement initiatives.
Keywords:
Methodology: The burn admission records were retrospectively interrogated for the years
Burn
2012–2013.
Developing world
Results: This audit covers an 18-month period in which 490 patients were admitted. Admitted days per percentage burn were 2.6 days per percentage total body surface area burnt. The mortality rate was 13%. Fourteen percent of patients met the criteria for referral to the provincial burn centre but for a variety of logistical reasons only 3% were transferred. Conclusion: We have redesigned the process of care without alteration of resources. Outcomes of burns less than 30% total body surface area are not acceptable which we believe reflects the lack of infrastructure and systems development. This audit has revealed a number of areas, which are suitable for dedicated quality improvement initiatives. # 2014 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
We know that burns are common in South Africa and affect the vulnerable sector of our population [1–3]. It is estimated that 3.2% of the South African population is burnt annually, 90% of those are minor burns but 0.2% (270 people per month) sustain a burn that needs specialised burn care and half of those need intensive care. Burns are the most common external cause of death in children less than 4 years old and mortality from burns in South Africa is 8.5 in 100 000 compared to the world average of 5 in 100 000 [4–6]. Many South African authors have described this high burden of injury [11–13] however little has been done to address the training of surgeons in the field of burns or towards the * Corresponding author. Tel.: +27 836557660; fax: +27 314670545. E-mail address: [email protected] (N.L. Allorto). http://dx.doi.org/10.1016/j.burns.2014.07.021 0305-4179/# 2014 Elsevier Ltd and ISBI. All rights reserved.
development of a sustainable and practical system of burn care. Burns are predominantly treated by general surgeons in Kwa-Zulu Natal, largely due to the greater number of general surgeons compared to plastic surgeons. Although expected to manage burns, there is not adequate training during registrar or fellowship years. Rotation through a burn unit or centre is not compulsory for trainees and the majority of qualified general surgeons have not had exposure to burn management. A single 1-hour seminar every two years on the management of burns is the only formal burn teaching in the surgical curriculum [3]. Even subspecialist trauma training has limited exposure to burn care. This results in a paucity of burn specialists and general surgeons with the appropriate skills to manage burns.
burns 41 (2015) 454–461
There is consensus in the developed world as to the definitions and levels of care for burn facilities, units and centres. A burn facility has burn beds that are in a shared environment with other surgical pathologies, has access to an operating theatre and onsite access to other services like paediatrics, radiology, pathology and a transfusion service. A burn unit treats burns of moderate severity, usually defined as percentage total body surface area (TBSA) between 15% and 40%, in a dedicated ward where children and adults are separated, with a temperature controlled theatre and provision of isolation cubicles. In addition psychological care as well as registered nurses with specific burn competencies are included in the standard of care for a burn unit. A burn centre includes all of the criteria for a burn unit and in addition has critical care services and should be associated with a trauma centre. The burn centre should manage severe burns that include large total body surface area (TBSA) burns, inhalation injuries and those requiring critical care. There is no formal systems approach to burns in South Africa. Resources such as burn units or centres are rare. This audit attempts to document the impact of restructuring the process of care in a single hospital on patient outcomes and to identify areas that are amenable to targeted quality improvement initiatives without the alteration of the resources or inputs.
2.
Setting
Pietermaritzburg is the capital of Kwa Zulu Natal (KZN) Province and is the largest city in the western part of the province with a population of 1 001 000 people. Western KZN is a predominantly rural province with a population of two million people, and consists of four health districts. Edendale Hospital is a regional hospital and part of the Pietermaritzburg metropolitan complex that includes a tertiary hospital and a district hospital. There are 900 beds to service general surgery, obstetrics, internal medicine, paediatrics and orthopaedic disciplines. The trauma service admits 150 patients per month [7] and 27 burn patients (15% of trauma admissions) with an additional burden of acute general surgical and elective cases. After hours there are 2 registrars and 2 interns on call to cover all general surgical emergencies including burns and trauma. The adult intensive care unit (ICU) has 6 ventilated and 3 high care beds. It is a closed unit receiving 93 referrals per month on average and 45 admissions. The paediatric ICU is also a closed unit having 5 beds and 25 admissions on average per month. Sixteen paediatric beds and 106 adults beds are allocated to general surgery. The burn ward has 8 beds that are interchangeable between adults and children. These beds are in 2 sections of 40 m2. There is no dedicated dressing room and only 1 registered nurse on duty per shift. We frequently have up to 45 admitted patients at a time, particularly in winter, and utilise surgical beds in the main ward to accommodate these patients.
3.
Systems
In Kwa Zulu Natal only one institution fulfils the criteria for a burn centre. All other hospitals managing burns only meet the
455
requirements for a burn facility despite the expectation to treat burns of moderate and major severity. Edendale Hospital is colloquially referred to as a burn unit and this reputation is almost certainly due to the development of a dedicated multidisciplinary team who manage all burn patients despite the lack of appropriate resources and infrastructure. Previously the general surgical service was responsible for management of burns [1,8]. Burn admissions were distributed among surgical firms managing a variety of surgical pathologies from appendicitis to breast cancer. Lack of motivation and limited theatre time led to a pervasive philosophy of wound bed preparation followed by delayed skin grafting resulting in prolonged hospital stays and although unquantified at the time, serious morbidity. Significant benefits to burned patients have been demonstrated in our country by improving burn care capability [9,10]. In response to this the Edendale Burn Service was instituted with the objective of providing a dedicated service for the improvement of this neglected field by changing process of care only.
4.
Restructuring the process of care
A number of processes of care were restructured to develop the Edendale burn service. A senior medical officer was employed to managed the burn service with part-time support from two specialist surgeons, one of whom is a registered trauma surgeon and the other completing sub-specialist training in critical care. A junior medical officer, a rotating general surgical registrar and an intern complete the team. The burns team manages all acute burns that are admitted during working-hours. After-hours the on call general surgical registrar is responsible for all burn admissions. The burn service has two theatre slates per week. The theatre is not temperature regulated. A dedicated anaesthetic registrar is assigned to the team on a quarterly basis with specialist support for every operating list. There is a weekly outpatient clinic for follow up visits. Edendale Hospital has no intensive care unit dedicated for burn care and patients with thermal injury requiring organ support, compete with trauma, medical, general surgery and obstetric patients for intensive care admission. Patients requiring complex plastic and reconstructive surgery are referred the tertiary hospital in Pietermaritzburg. Major acute burns are referred to the Albert Luthuli Burn Centre in Durban once they have been appropriately resuscitated. The criteria for transfer to the Burn Centre are a burn of total body surface area (TBSA) over 35%; full thickness burns in special areas, children under one year old with full thickness burns and all children with full thickness burns of more than 5% TBSA, inhalation injury and need for intensive care. A multi-disciplinary team meeting is held every Monday afternoon and attended by the anaesthetist, dietician, occupational therapist, physiotherapist, nursing staff and surgical burn team. Each admitted patient is discussed in terms of progress and holistic needs and the operative plan for each patient is defined. Unexpected deaths and expected deaths are discussed in this forum as both a quality improvement structure and a means of debriefing. Mortality and morbidity is also discussed monthly in the general surgical forum, as all
456
burns 41 (2015) 454–461
members of the surgical department are involved with acute management of burn patients. A burn specific sheet (Figs. 1–3) has been introduced for all burn admissions. This sheet includes epidemiological data, details of the circumstances at the time of the burn as well as guidance through assessment of airway risk criteria, extent of the burn using the Lund and Browder charts and fluid resuscitation. The degree of compliance with all the data fields is dependant on the discretion of the doctor. Most burn admissions are discussed with a burn-experienced senior doctor. A burn specific operative record sheet has similarly been designed. The admission, operative and discharge data was then captured onto an access based electronic programme.
5.
Methodology
Data from the admission/discharge and operative record was collated and reviewed by the primary author. The collected data included demographic details, extent and mechanism of the burns as well as data pertaining to operative interventions and mortality.
6.
adult intensive care unit and 27 (10% of total burn admissions) paediatric admissions. The average TBSA for ICU admission was 12% (range 4–35%). The mortality rate was 13% in this cohort of patients. Mortalities are illustrated in Fig. 5 and Table 1. There were 64 deaths of which 74% were unexpected. All of the 12 children who died were classified as unexpected and potentially preventable. The cause of death in these children was acute respiratory compromise or acute severe sepsis. In adult patients 35% were expected deaths with an average TBSA of 54% and Baux score of 88. The remaining 65% of adult deaths were unanticipated with an average TBSA of 18% and Baux score of 66. Nine patients (50%) in the expected death group had inhalational injury compared to five (15%) in the unexpected death group. Sixty-seven patients in this cohort met the criteria for referral to the burn centre, however only three were accepted for transfer. The remainder were treated at our institution and the mortality in this group was 65% compared to 13% overall mortality. The criteria met for transfer to the quaternary centre at Albert Luthuli Hospital in our patients was percentage surface over 35% TBSA in 37 patients, need for intensive care and ventilation in 13, burn size and need for ICU in seven, and full thickness burns in special areas in seven patients.
Results 7.
The series reviews admissions from January 2012 to June 2013 with a total of 490 patient admissions. Children account for 48% of admissions and two thirds of the children admitted were less than five years old. Two hundred and eighty seven of the patients were male (59%). Flame is the predominant aetiology in adult admissions (70%) and scalds the most common cause in children (73%). Chemical, electrical and contact burns account for the remaining aetiologies. Comorbidities were documented in 10% of patients, which included epilepsy (14), alcohol abuse (11), hypertension (9), psychiatric disease (6), asthma (6), diabetes mellitus (3), cerebral palsy (1), and mental retardation (1). The average percentage surface area in adults was 13% with a large range from eight to 77%. In children the average was 9% (range 1–50%). Length of hospital stay ranged from one to 137 days with an average of 28 days for survivors. This equated to 2.6 days per percentage TBSA burn on average. In the group that died average length of stay was 3 days in the expected death group and 32 days in the unexpected group. Three hundred and fifty eight operative procedures were booked during this period with 38 (11%) cancellations. There were 76 (24%) dressing changes in theatre and 244 (76%) operative interventions. In only 48 (20%) patients did we manage to perform early surgery. Twenty-four of these patients had single stage excision and grafting and the other 24 patients had early initial debridement but delayed further excision and or grafting. Theatre utilisation is illustration in Fig. 4. Fifteen percent of patients operated on had more than one operation, with the average being two operations and a range from two to five. The majority (86%) of patients are managed without operative management of the burn wound. Over this period there were 24 (11% of the total burn admissions and 2% of all intensive care admissions) to the
Discussion
A health care system consists of resources or inputs, which must be utilised by a process of care. The inputs and processes of care interact to produce an outcome. Outcomes are dependant on both components of the system and may be affected by altering either or both of these components. Realizing that resource constraints could not easily be addressed in our institution we aimed at improving efficiency in the system by refining process of care. We have developed a dedicated burn service which functions with a team of doctors, who exclusively manage burns during working hours, as well as a multi-disciplinary component, but without the benefit of the physical infrastructure of a burn unit. The service is designed as a concept of care and functions with eight high care beds, with the majority of admissions going to the general surgical wards and lack of dedicated cover after hours. We do not fulfil criteria to be a burn unit and would be classified as a burn facility only. However the current data shows that we manage large volumes of moderately severe burns. Fourteen percent of the patients we admit would be more appropriately managed in a burn centre. Our average TBSA is 9% in children and 13% in adults with 27 admissions per month compared to 18% average TBSA and 19 admissions per month in the referral burn centre [3]. Despite development of a dedicated burn team the duration of hospital stay is significant for a cohort of moderate severity burns. We have attempted shift away from awaiting spontaneous eschar separation prior to performing delayed skin grafting to that of a more contemporary wound management strategy. This includes modern dressings, for example nano-crystalline silver and hydrofibres, in conjunction with attempted early excision and skin grafting. Although
burns 41 (2015) 454–461
Fig. 1 – Injury and patient details of Admission Form.
457
458
burns 41 (2015) 454–461
Fig. 2 – Burn assessment of Admission Form.
early excision of deep wounds and immediate skin grafting is planned, this strategy still poses logistical difficulties and we are often constrained to a strategy of staged excision and grafting. This is excision of 10–15% TBSA every three days till elimination of all eschar and then skin grafting till complete
wound closure. The limited availability of theatre time, blood products and post-operative intensive care beds has necessitated this approach. There is published data on the deficiency of ICU beds in the public sector in the South Africa which reports a ratio of 1
burns 41 (2015) 454–461
459
Fig. 3 – Fluid resuscitation tool and analgesia guidelines of Admission Form.
hospital bed per 32 000 of population and with only of 3.3% of all available hospital beds being ICU beds. In KZN it is estimated that 39 ICU beds are not utilised due to staff shortages [17]. This national shortage of ICU beds impacts on the critically ill burn population. The majority of ICU admissions for burns at our institution are for upper airway inhalational injuries or large TBSA burns in children.
Paediatric ICU has no limits to admission of burn patients but the adult unit will only consider burns of less than 30% TBSA for admission. This policy is justified by the dismal prognosis of major burns in our institution. Failure to get to the appropriate level of care remains an obstacle and only 4% (3 out of 67) of patients who met the referral criteria for transfer to the Albert Luthuli burn centre were actually transferred.
460
burns 41 (2015) 454–461
Fig. 4 – Utilisation of theatre slates.
This represents system failure on a large scale. The single burn centre has recently published data on the spectrum of burns in their unit with an average TBSA burn of 18% [3]. This is an inappropriate severity of burn for such a centre to be treating and reflects an overwhelmed system where inadequate secondary level burn care results in patients with minor to moderate burns being inappropriately managed by the burn centre. The development of appropriate burn services and units in the province would relieve the burn centre and allow it to focus on caring for severe burns of large TBSA, inhalational burns and those who require intensive care. There is an ongoing conversation amongst surgeons caring for burns in our province regarding system development and appropriate training of surgeons to address this issue. This audit has helped us to identify a number of areas in our institution, which can be targeted by specific quality improvement programmes. There were no analgesia protocols or strategies in place for dressing changes. This in combination with a lack of dedicated dressing area resulted in the majority of the major burn dressing changes being performed in the operating room and accounting for a quarter of booked theatre cases. This in turn impacts our ability to perform early burn wound excision and affects duration of hospital stay. The low rate of operative intervention (14%) is probably due to this inefficient use of resources and development of a dedicated dressing room would allow for more valuable use of the available operating time. Building capability to perform early excision and grafting where needed is also likely to contribute to improving our mortality rate. Inhalational injuries are not routinely assessed by bronchoscopy in our service and the diagnosis of these injuries remains a clinical one. We do have access to bronchoscopy and need to utilise it on a more frequent basis [16]. This is another potential target for a quality improvement programmes and a protocol that formalizes the
Fig. 5 – Expected versus unexpected deaths.
endoscopic assessment and staging of inhalation injury is being introduced. We define an unexpected death as one that is associated with avoidable factors. We use a peer review process [13] at the monthly mortality meetings to elucidate and remediate these factors. In our institution burn surface areas greater than 40% are fatal [1]. Admissions to our institution of burns greater than 40% are referred to the burn centre and if transfer is not accepted we shift the focus of care from therapy to palliation. Deaths in this cohort of patients are deemed to be expected mortalities and the average TBSA in the expected death group was 54%. The mortality rate is however high relative to the average severity of burns admitted and we are currently reviewing all burn mortalities in our service. The inefficient use of theatre time and the failure to achieve early excision and grafting is one explanation for this high mortality rate. Other burn units in the country also describe high mortality rates [14,15] and this reflects the inadequate care packages for burns in South Africa. In our service lack of access to isolation cubicles and mechanical ventilation, inadequate nursing staff and after hours medical care, limited access to the operating room and inconsistent access to advanced dressings all contribute to the high mortality rate. However the number of unexpected deaths with only moderate (18%) TBSA burns are a major concern and burn deaths will undergo more detailed audit. Record keeping is a major deficit to be addressed. Paper documentation is prone to the omission of data. Subsequently we have been unable to comment on depth of burn and Abbreviated Burn Severity Index in this cohort. To tackle this deficiency we are developing a hybrid electronic medical record and registry tool to improve both the quality of the medical record keeping and the collection of data for quality improvement purposes.
Table 1 – Comparison of mortalities.
Number Average age (range) Average % TBSA (range) Average Baux score (range)
Adult expected
Adult unexpected
Children
18 34 (18–70) 54 (40–70) 88 (54–110)
34 48 (14–77) 18 (9–46) 66 (52–103)
12 26 months (8 months–5 years) Not applicable Not applicable
burns 41 (2015) 454–461
8.
Conclusion
The development of a dedicated burn service in a single institution represents an attempt to alter the process of care without increased resource allocation. Burns patients in our institution are now treated by a multidisciplinary team, which provides holistic management. However lack of appropriate infrastructure does seem to influence the length of hospital stay and high mortality rate. Without a significant investment in infrastructure such as a dedicated burns intensive care unit, temperature controlled readily available operating theatres, isolation cubicles and more nursing staff it is unlikely that we will be able to improve survival rates for large burns in our institution and duration of hospital stay for the moderate severity burns. The improvement of systems on a provincial scale with the implementation of appropriate referral mechanisms and the formalisation of registrar and fellow training in the management of burns is needed. We have however identified a number of areas, which can feasibly be targeted by multifaceted quality improvement programmes within our institution. These include a closer analysis and audit of all unexpected mortalities, a policy to increase the use of bronchoscopy to accurately stage inhalational burns, the creation of a dedicated dressing area and analgesia/sedation protocols with more efficient use of theatre time targeting early excision and grafting and the implementation of a robust electronic medical record system for improved data collection and audit.
[4] [5]
[6]
[7]
[8]
[9]
[10] [11] [12]
[13]
Conflict of interest None declared.
[14]
references [15] [1] Allorto NL, Oosthuizen GV, Clarke DL, Muckart DJ. The spectrum and outcome of burns in a regional hospital in South Africa. Burns 2009;35(7):1004–8. http://dx.doi.org/ 10.1016/j.burns.01.004. [2] Gevaart-Durkin A, Swart D, Chowdhury Z. A study of energy-related injuries from hospital admissions among children and adults in South Africa. Burns 2014. http:// dx.doi.org/10.1016/j.burns.2013.12.014. [3] Den Hollander D, Albert M, Strand A, Hardcastle TC. Epidemiology and referral patterns of burns admitted to the
[16]
[17]
461
Burns Centre at Inkosi Albert Luthuli Central Hospital, Durban. Burns 2014. http://dx.doi.org/10.1016/ j.burns.2013.12.018. Integrated strategic framework for the prevention of injury and violence in South Africa; 2012. Strategic framework for the modernisation of tertiary services. South African National Department of Health; 2003. A profile of fatal injuries in South Africa: 10th annual report of the National Injury Mortality Surveillance System; 2008. Laing GL, Skinner DL, Bruce JL, Aldous C, Oosthuizen GV, Clarke DL. Understanding the burden outcome of trauma care drives a new trauma systems model. World J Surg 2014;38:1699–706. Allorto NL, Clarke DL, Thomson SR. A cost model case comparison of current versus modern management of burns at a regional hospital in South Africa. Burns 2011;37(6):1033–7. http://dx.doi.org/10.1016/ j.burns.2011.04.004. Bauling PC, Kruger JS, van der Vyver JD, van Rooyen M, Lombard JH. Early benefits of improved burn-care capability at Kalafong Hospital. S Afr Med J 1992;81:310–2. Eyal AS, Kemp M, Luvhengo T. A 10-year audit of burns at Kalafong Hospital. Burns 2007;33:393–5. Rode H, Berg AM, Rogers A. Burn care in South Africa. Ann Burns Fire Disasters 2011;24(1):7–8. Wesson HK, Bachani AM, Mtambeka P, Schulman D, Mavengere C, Stevens KA, et al. Pediatric burn injuries in South Africa: a 15-year analysis of hospital data. Injury 2013;44(11):77–82. http://dx.doi.org/10.1016/ j.injury.2012.12.017. Clarke DL, Furlong H, Laing GL, Aldous C, Thomson SR. Using a structured morbidity and mortality meeting to understand the contribution of human error to adverse surgical events in a South African regional hospital. S Afr J Surg 2013;51(4):122–6. http://dx.doi.org/10.7196/sajs.1537. Scheven D, Barker P, Govindasamy J. Burns in rural KwaZulu Natal: epidemiology and the need for community health education. Burns 2012;38(8):1224–30. http:// dx.doi.org/10.1016/j.burns.2012.04.001. Maritz D, Wallis L, Van Der Merwe E, Nel D. The aetiology of adult burns in the Western Cape, South Africa. Burns 2012;38(1):120–7. http://dx.doi.org/10.1016/ j.burns.2010.12.007. Marek K, Piotr W, Stanisław S, Stefan G, Justyna G, Mariusz N, et al. Fibreoptic bronchoscopy in routine clinical practice in confirming the diagnosis and treatment of inhalation burns. Burns 2007;33(5):554–60. Bhagwanjee S, Scribante J. National audit of critical care resources in South Africa – unit and bed distribution. S Afr Med J 2007;97(12 (Pt 3)):1311–4.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Merits and challenges in the development of a dedicated burn service at a regional hospital in South Africa Nikki L. Allorto *, Damian L. Clarke Department of Surgery, Pietermaritzburg Metropolitan Complex, University of Kwa-Zulu Natal, South Africa
article info
abstract
Article history:
Introduction: The Edendale Hospital Burn Service was initiated in 2011 to improve the quality of
Accepted 28 July 2014
burn care at a regional hospital. This audit reviews the merits and challenges in developing such a service and identifies areas on which to focus quality improvement initiatives.
Keywords:
Methodology: The burn admission records were retrospectively interrogated for the years
Burn
2012–2013.
Developing world
Results: This audit covers an 18-month period in which 490 patients were admitted. Admitted days per percentage burn were 2.6 days per percentage total body surface area burnt. The mortality rate was 13%. Fourteen percent of patients met the criteria for referral to the provincial burn centre but for a variety of logistical reasons only 3% were transferred. Conclusion: We have redesigned the process of care without alteration of resources. Outcomes of burns less than 30% total body surface area are not acceptable which we believe reflects the lack of infrastructure and systems development. This audit has revealed a number of areas, which are suitable for dedicated quality improvement initiatives. # 2014 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
We know that burns are common in South Africa and affect the vulnerable sector of our population [1–3]. It is estimated that 3.2% of the South African population is burnt annually, 90% of those are minor burns but 0.2% (270 people per month) sustain a burn that needs specialised burn care and half of those need intensive care. Burns are the most common external cause of death in children less than 4 years old and mortality from burns in South Africa is 8.5 in 100 000 compared to the world average of 5 in 100 000 [4–6]. Many South African authors have described this high burden of injury [11–13] however little has been done to address the training of surgeons in the field of burns or towards the * Corresponding author. Tel.: +27 836557660; fax: +27 314670545. E-mail address: [email protected] (N.L. Allorto). http://dx.doi.org/10.1016/j.burns.2014.07.021 0305-4179/# 2014 Elsevier Ltd and ISBI. All rights reserved.
development of a sustainable and practical system of burn care. Burns are predominantly treated by general surgeons in Kwa-Zulu Natal, largely due to the greater number of general surgeons compared to plastic surgeons. Although expected to manage burns, there is not adequate training during registrar or fellowship years. Rotation through a burn unit or centre is not compulsory for trainees and the majority of qualified general surgeons have not had exposure to burn management. A single 1-hour seminar every two years on the management of burns is the only formal burn teaching in the surgical curriculum [3]. Even subspecialist trauma training has limited exposure to burn care. This results in a paucity of burn specialists and general surgeons with the appropriate skills to manage burns.
burns 41 (2015) 454–461
There is consensus in the developed world as to the definitions and levels of care for burn facilities, units and centres. A burn facility has burn beds that are in a shared environment with other surgical pathologies, has access to an operating theatre and onsite access to other services like paediatrics, radiology, pathology and a transfusion service. A burn unit treats burns of moderate severity, usually defined as percentage total body surface area (TBSA) between 15% and 40%, in a dedicated ward where children and adults are separated, with a temperature controlled theatre and provision of isolation cubicles. In addition psychological care as well as registered nurses with specific burn competencies are included in the standard of care for a burn unit. A burn centre includes all of the criteria for a burn unit and in addition has critical care services and should be associated with a trauma centre. The burn centre should manage severe burns that include large total body surface area (TBSA) burns, inhalation injuries and those requiring critical care. There is no formal systems approach to burns in South Africa. Resources such as burn units or centres are rare. This audit attempts to document the impact of restructuring the process of care in a single hospital on patient outcomes and to identify areas that are amenable to targeted quality improvement initiatives without the alteration of the resources or inputs.
2.
Setting
Pietermaritzburg is the capital of Kwa Zulu Natal (KZN) Province and is the largest city in the western part of the province with a population of 1 001 000 people. Western KZN is a predominantly rural province with a population of two million people, and consists of four health districts. Edendale Hospital is a regional hospital and part of the Pietermaritzburg metropolitan complex that includes a tertiary hospital and a district hospital. There are 900 beds to service general surgery, obstetrics, internal medicine, paediatrics and orthopaedic disciplines. The trauma service admits 150 patients per month [7] and 27 burn patients (15% of trauma admissions) with an additional burden of acute general surgical and elective cases. After hours there are 2 registrars and 2 interns on call to cover all general surgical emergencies including burns and trauma. The adult intensive care unit (ICU) has 6 ventilated and 3 high care beds. It is a closed unit receiving 93 referrals per month on average and 45 admissions. The paediatric ICU is also a closed unit having 5 beds and 25 admissions on average per month. Sixteen paediatric beds and 106 adults beds are allocated to general surgery. The burn ward has 8 beds that are interchangeable between adults and children. These beds are in 2 sections of 40 m2. There is no dedicated dressing room and only 1 registered nurse on duty per shift. We frequently have up to 45 admitted patients at a time, particularly in winter, and utilise surgical beds in the main ward to accommodate these patients.
3.
Systems
In Kwa Zulu Natal only one institution fulfils the criteria for a burn centre. All other hospitals managing burns only meet the
455
requirements for a burn facility despite the expectation to treat burns of moderate and major severity. Edendale Hospital is colloquially referred to as a burn unit and this reputation is almost certainly due to the development of a dedicated multidisciplinary team who manage all burn patients despite the lack of appropriate resources and infrastructure. Previously the general surgical service was responsible for management of burns [1,8]. Burn admissions were distributed among surgical firms managing a variety of surgical pathologies from appendicitis to breast cancer. Lack of motivation and limited theatre time led to a pervasive philosophy of wound bed preparation followed by delayed skin grafting resulting in prolonged hospital stays and although unquantified at the time, serious morbidity. Significant benefits to burned patients have been demonstrated in our country by improving burn care capability [9,10]. In response to this the Edendale Burn Service was instituted with the objective of providing a dedicated service for the improvement of this neglected field by changing process of care only.
4.
Restructuring the process of care
A number of processes of care were restructured to develop the Edendale burn service. A senior medical officer was employed to managed the burn service with part-time support from two specialist surgeons, one of whom is a registered trauma surgeon and the other completing sub-specialist training in critical care. A junior medical officer, a rotating general surgical registrar and an intern complete the team. The burns team manages all acute burns that are admitted during working-hours. After-hours the on call general surgical registrar is responsible for all burn admissions. The burn service has two theatre slates per week. The theatre is not temperature regulated. A dedicated anaesthetic registrar is assigned to the team on a quarterly basis with specialist support for every operating list. There is a weekly outpatient clinic for follow up visits. Edendale Hospital has no intensive care unit dedicated for burn care and patients with thermal injury requiring organ support, compete with trauma, medical, general surgery and obstetric patients for intensive care admission. Patients requiring complex plastic and reconstructive surgery are referred the tertiary hospital in Pietermaritzburg. Major acute burns are referred to the Albert Luthuli Burn Centre in Durban once they have been appropriately resuscitated. The criteria for transfer to the Burn Centre are a burn of total body surface area (TBSA) over 35%; full thickness burns in special areas, children under one year old with full thickness burns and all children with full thickness burns of more than 5% TBSA, inhalation injury and need for intensive care. A multi-disciplinary team meeting is held every Monday afternoon and attended by the anaesthetist, dietician, occupational therapist, physiotherapist, nursing staff and surgical burn team. Each admitted patient is discussed in terms of progress and holistic needs and the operative plan for each patient is defined. Unexpected deaths and expected deaths are discussed in this forum as both a quality improvement structure and a means of debriefing. Mortality and morbidity is also discussed monthly in the general surgical forum, as all
456
burns 41 (2015) 454–461
members of the surgical department are involved with acute management of burn patients. A burn specific sheet (Figs. 1–3) has been introduced for all burn admissions. This sheet includes epidemiological data, details of the circumstances at the time of the burn as well as guidance through assessment of airway risk criteria, extent of the burn using the Lund and Browder charts and fluid resuscitation. The degree of compliance with all the data fields is dependant on the discretion of the doctor. Most burn admissions are discussed with a burn-experienced senior doctor. A burn specific operative record sheet has similarly been designed. The admission, operative and discharge data was then captured onto an access based electronic programme.
5.
Methodology
Data from the admission/discharge and operative record was collated and reviewed by the primary author. The collected data included demographic details, extent and mechanism of the burns as well as data pertaining to operative interventions and mortality.
6.
adult intensive care unit and 27 (10% of total burn admissions) paediatric admissions. The average TBSA for ICU admission was 12% (range 4–35%). The mortality rate was 13% in this cohort of patients. Mortalities are illustrated in Fig. 5 and Table 1. There were 64 deaths of which 74% were unexpected. All of the 12 children who died were classified as unexpected and potentially preventable. The cause of death in these children was acute respiratory compromise or acute severe sepsis. In adult patients 35% were expected deaths with an average TBSA of 54% and Baux score of 88. The remaining 65% of adult deaths were unanticipated with an average TBSA of 18% and Baux score of 66. Nine patients (50%) in the expected death group had inhalational injury compared to five (15%) in the unexpected death group. Sixty-seven patients in this cohort met the criteria for referral to the burn centre, however only three were accepted for transfer. The remainder were treated at our institution and the mortality in this group was 65% compared to 13% overall mortality. The criteria met for transfer to the quaternary centre at Albert Luthuli Hospital in our patients was percentage surface over 35% TBSA in 37 patients, need for intensive care and ventilation in 13, burn size and need for ICU in seven, and full thickness burns in special areas in seven patients.
Results 7.
The series reviews admissions from January 2012 to June 2013 with a total of 490 patient admissions. Children account for 48% of admissions and two thirds of the children admitted were less than five years old. Two hundred and eighty seven of the patients were male (59%). Flame is the predominant aetiology in adult admissions (70%) and scalds the most common cause in children (73%). Chemical, electrical and contact burns account for the remaining aetiologies. Comorbidities were documented in 10% of patients, which included epilepsy (14), alcohol abuse (11), hypertension (9), psychiatric disease (6), asthma (6), diabetes mellitus (3), cerebral palsy (1), and mental retardation (1). The average percentage surface area in adults was 13% with a large range from eight to 77%. In children the average was 9% (range 1–50%). Length of hospital stay ranged from one to 137 days with an average of 28 days for survivors. This equated to 2.6 days per percentage TBSA burn on average. In the group that died average length of stay was 3 days in the expected death group and 32 days in the unexpected group. Three hundred and fifty eight operative procedures were booked during this period with 38 (11%) cancellations. There were 76 (24%) dressing changes in theatre and 244 (76%) operative interventions. In only 48 (20%) patients did we manage to perform early surgery. Twenty-four of these patients had single stage excision and grafting and the other 24 patients had early initial debridement but delayed further excision and or grafting. Theatre utilisation is illustration in Fig. 4. Fifteen percent of patients operated on had more than one operation, with the average being two operations and a range from two to five. The majority (86%) of patients are managed without operative management of the burn wound. Over this period there were 24 (11% of the total burn admissions and 2% of all intensive care admissions) to the
Discussion
A health care system consists of resources or inputs, which must be utilised by a process of care. The inputs and processes of care interact to produce an outcome. Outcomes are dependant on both components of the system and may be affected by altering either or both of these components. Realizing that resource constraints could not easily be addressed in our institution we aimed at improving efficiency in the system by refining process of care. We have developed a dedicated burn service which functions with a team of doctors, who exclusively manage burns during working hours, as well as a multi-disciplinary component, but without the benefit of the physical infrastructure of a burn unit. The service is designed as a concept of care and functions with eight high care beds, with the majority of admissions going to the general surgical wards and lack of dedicated cover after hours. We do not fulfil criteria to be a burn unit and would be classified as a burn facility only. However the current data shows that we manage large volumes of moderately severe burns. Fourteen percent of the patients we admit would be more appropriately managed in a burn centre. Our average TBSA is 9% in children and 13% in adults with 27 admissions per month compared to 18% average TBSA and 19 admissions per month in the referral burn centre [3]. Despite development of a dedicated burn team the duration of hospital stay is significant for a cohort of moderate severity burns. We have attempted shift away from awaiting spontaneous eschar separation prior to performing delayed skin grafting to that of a more contemporary wound management strategy. This includes modern dressings, for example nano-crystalline silver and hydrofibres, in conjunction with attempted early excision and skin grafting. Although
burns 41 (2015) 454–461
Fig. 1 – Injury and patient details of Admission Form.
457
458
burns 41 (2015) 454–461
Fig. 2 – Burn assessment of Admission Form.
early excision of deep wounds and immediate skin grafting is planned, this strategy still poses logistical difficulties and we are often constrained to a strategy of staged excision and grafting. This is excision of 10–15% TBSA every three days till elimination of all eschar and then skin grafting till complete
wound closure. The limited availability of theatre time, blood products and post-operative intensive care beds has necessitated this approach. There is published data on the deficiency of ICU beds in the public sector in the South Africa which reports a ratio of 1
burns 41 (2015) 454–461
459
Fig. 3 – Fluid resuscitation tool and analgesia guidelines of Admission Form.
hospital bed per 32 000 of population and with only of 3.3% of all available hospital beds being ICU beds. In KZN it is estimated that 39 ICU beds are not utilised due to staff shortages [17]. This national shortage of ICU beds impacts on the critically ill burn population. The majority of ICU admissions for burns at our institution are for upper airway inhalational injuries or large TBSA burns in children.
Paediatric ICU has no limits to admission of burn patients but the adult unit will only consider burns of less than 30% TBSA for admission. This policy is justified by the dismal prognosis of major burns in our institution. Failure to get to the appropriate level of care remains an obstacle and only 4% (3 out of 67) of patients who met the referral criteria for transfer to the Albert Luthuli burn centre were actually transferred.
460
burns 41 (2015) 454–461
Fig. 4 – Utilisation of theatre slates.
This represents system failure on a large scale. The single burn centre has recently published data on the spectrum of burns in their unit with an average TBSA burn of 18% [3]. This is an inappropriate severity of burn for such a centre to be treating and reflects an overwhelmed system where inadequate secondary level burn care results in patients with minor to moderate burns being inappropriately managed by the burn centre. The development of appropriate burn services and units in the province would relieve the burn centre and allow it to focus on caring for severe burns of large TBSA, inhalational burns and those who require intensive care. There is an ongoing conversation amongst surgeons caring for burns in our province regarding system development and appropriate training of surgeons to address this issue. This audit has helped us to identify a number of areas in our institution, which can be targeted by specific quality improvement programmes. There were no analgesia protocols or strategies in place for dressing changes. This in combination with a lack of dedicated dressing area resulted in the majority of the major burn dressing changes being performed in the operating room and accounting for a quarter of booked theatre cases. This in turn impacts our ability to perform early burn wound excision and affects duration of hospital stay. The low rate of operative intervention (14%) is probably due to this inefficient use of resources and development of a dedicated dressing room would allow for more valuable use of the available operating time. Building capability to perform early excision and grafting where needed is also likely to contribute to improving our mortality rate. Inhalational injuries are not routinely assessed by bronchoscopy in our service and the diagnosis of these injuries remains a clinical one. We do have access to bronchoscopy and need to utilise it on a more frequent basis [16]. This is another potential target for a quality improvement programmes and a protocol that formalizes the
Fig. 5 – Expected versus unexpected deaths.
endoscopic assessment and staging of inhalation injury is being introduced. We define an unexpected death as one that is associated with avoidable factors. We use a peer review process [13] at the monthly mortality meetings to elucidate and remediate these factors. In our institution burn surface areas greater than 40% are fatal [1]. Admissions to our institution of burns greater than 40% are referred to the burn centre and if transfer is not accepted we shift the focus of care from therapy to palliation. Deaths in this cohort of patients are deemed to be expected mortalities and the average TBSA in the expected death group was 54%. The mortality rate is however high relative to the average severity of burns admitted and we are currently reviewing all burn mortalities in our service. The inefficient use of theatre time and the failure to achieve early excision and grafting is one explanation for this high mortality rate. Other burn units in the country also describe high mortality rates [14,15] and this reflects the inadequate care packages for burns in South Africa. In our service lack of access to isolation cubicles and mechanical ventilation, inadequate nursing staff and after hours medical care, limited access to the operating room and inconsistent access to advanced dressings all contribute to the high mortality rate. However the number of unexpected deaths with only moderate (18%) TBSA burns are a major concern and burn deaths will undergo more detailed audit. Record keeping is a major deficit to be addressed. Paper documentation is prone to the omission of data. Subsequently we have been unable to comment on depth of burn and Abbreviated Burn Severity Index in this cohort. To tackle this deficiency we are developing a hybrid electronic medical record and registry tool to improve both the quality of the medical record keeping and the collection of data for quality improvement purposes.
Table 1 – Comparison of mortalities.
Number Average age (range) Average % TBSA (range) Average Baux score (range)
Adult expected
Adult unexpected
Children
18 34 (18–70) 54 (40–70) 88 (54–110)
34 48 (14–77) 18 (9–46) 66 (52–103)
12 26 months (8 months–5 years) Not applicable Not applicable
burns 41 (2015) 454–461
8.
Conclusion
The development of a dedicated burn service in a single institution represents an attempt to alter the process of care without increased resource allocation. Burns patients in our institution are now treated by a multidisciplinary team, which provides holistic management. However lack of appropriate infrastructure does seem to influence the length of hospital stay and high mortality rate. Without a significant investment in infrastructure such as a dedicated burns intensive care unit, temperature controlled readily available operating theatres, isolation cubicles and more nursing staff it is unlikely that we will be able to improve survival rates for large burns in our institution and duration of hospital stay for the moderate severity burns. The improvement of systems on a provincial scale with the implementation of appropriate referral mechanisms and the formalisation of registrar and fellow training in the management of burns is needed. We have however identified a number of areas, which can feasibly be targeted by multifaceted quality improvement programmes within our institution. These include a closer analysis and audit of all unexpected mortalities, a policy to increase the use of bronchoscopy to accurately stage inhalational burns, the creation of a dedicated dressing area and analgesia/sedation protocols with more efficient use of theatre time targeting early excision and grafting and the implementation of a robust electronic medical record system for improved data collection and audit.
[4] [5]
[6]
[7]
[8]
[9]
[10] [11] [12]
[13]
Conflict of interest None declared.
[14]
references [15] [1] Allorto NL, Oosthuizen GV, Clarke DL, Muckart DJ. The spectrum and outcome of burns in a regional hospital in South Africa. Burns 2009;35(7):1004–8. http://dx.doi.org/ 10.1016/j.burns.01.004. [2] Gevaart-Durkin A, Swart D, Chowdhury Z. A study of energy-related injuries from hospital admissions among children and adults in South Africa. Burns 2014. http:// dx.doi.org/10.1016/j.burns.2013.12.014. [3] Den Hollander D, Albert M, Strand A, Hardcastle TC. Epidemiology and referral patterns of burns admitted to the
[16]
[17]
461
Burns Centre at Inkosi Albert Luthuli Central Hospital, Durban. Burns 2014. http://dx.doi.org/10.1016/ j.burns.2013.12.018. Integrated strategic framework for the prevention of injury and violence in South Africa; 2012. Strategic framework for the modernisation of tertiary services. South African National Department of Health; 2003. A profile of fatal injuries in South Africa: 10th annual report of the National Injury Mortality Surveillance System; 2008. Laing GL, Skinner DL, Bruce JL, Aldous C, Oosthuizen GV, Clarke DL. Understanding the burden outcome of trauma care drives a new trauma systems model. World J Surg 2014;38:1699–706. Allorto NL, Clarke DL, Thomson SR. A cost model case comparison of current versus modern management of burns at a regional hospital in South Africa. Burns 2011;37(6):1033–7. http://dx.doi.org/10.1016/ j.burns.2011.04.004. Bauling PC, Kruger JS, van der Vyver JD, van Rooyen M, Lombard JH. Early benefits of improved burn-care capability at Kalafong Hospital. S Afr Med J 1992;81:310–2. Eyal AS, Kemp M, Luvhengo T. A 10-year audit of burns at Kalafong Hospital. Burns 2007;33:393–5. Rode H, Berg AM, Rogers A. Burn care in South Africa. Ann Burns Fire Disasters 2011;24(1):7–8. Wesson HK, Bachani AM, Mtambeka P, Schulman D, Mavengere C, Stevens KA, et al. Pediatric burn injuries in South Africa: a 15-year analysis of hospital data. Injury 2013;44(11):77–82. http://dx.doi.org/10.1016/ j.injury.2012.12.017. Clarke DL, Furlong H, Laing GL, Aldous C, Thomson SR. Using a structured morbidity and mortality meeting to understand the contribution of human error to adverse surgical events in a South African regional hospital. S Afr J Surg 2013;51(4):122–6. http://dx.doi.org/10.7196/sajs.1537. Scheven D, Barker P, Govindasamy J. Burns in rural KwaZulu Natal: epidemiology and the need for community health education. Burns 2012;38(8):1224–30. http:// dx.doi.org/10.1016/j.burns.2012.04.001. Maritz D, Wallis L, Van Der Merwe E, Nel D. The aetiology of adult burns in the Western Cape, South Africa. Burns 2012;38(1):120–7. http://dx.doi.org/10.1016/ j.burns.2010.12.007. Marek K, Piotr W, Stanisław S, Stefan G, Justyna G, Mariusz N, et al. Fibreoptic bronchoscopy in routine clinical practice in confirming the diagnosis and treatment of inhalation burns. Burns 2007;33(5):554–60. Bhagwanjee S, Scribante J. National audit of critical care resources in South Africa – unit and bed distribution. S Afr Med J 2007;97(12 (Pt 3)):1311–4.
