Susan E Adams1, Andrew Holland2, Julie Brown3. 1. Department Paediatric Surgery, Sydney Children's Hospital, Randwick, NSW, Australia; School of Women's and Children's Health, University of NSW, Kensington, NSW, 2033, Australia; Neuroscience Research Australia (NeuRA), Randwick, NSW, 2031, Australia; School of Medical Science, University of NSW, Kensington, NSW, 2033, Australia. Electronic address: susan.adams@unsw.edu.au. 2. Department of Academic Surgery, Royal Alexandria Hospital for Children, Westmead, NSW, 2145, Australia; Discipline of Child and Adolescent Health, Sydney Medical School, University of Sydney, NSW, 2006, Australia. 3. Neuroscience Research Australia (NeuRA), Randwick, NSW, 2031, Australia; School of Medical Science, University of NSW, Kensington, NSW, 2033, Australia.
Abstract
Since the 1980's, paediatric surgeons have increasingly managed blunt splenic injury (BSI) in children non-operatively. However, studies in North America have shown higher operation rates in non-paediatric centres and by adult surgeons. This association has not been examined elsewhere. OBJECTIVE: To investigate the management of BSI in New South Wales (NSW) children, to determine the patient and hospital factors related to the odds of operation. Secondarily, to investigate whether the likelihood of operation varied by year. METHODS: Children age 0-16 admitted to a NSW hospital between July 2000 and December 2011 with a diagnosis of BSI were identified in the NSW Admitted Patient Data Collection, and linked to deaths data from Registry of Births Deaths and Marriages, and Bureau of Statistics. The operation rate was calculated and compared between different hospital types. Univariable analysis was used to determine patient and hospital factors associated with operative management. The difference in the odds of operation between the oldest data (July 2000-December 2005) and most recent (January 2006-December 2011) was also examined. Multivariable logistic regression with stepwise elimination was then performed to determine likelihood of operative management according to hospital category and era, adjusting for potential confounders. RESULTS: 955 cases were identified, with 101(10.6%) managed operatively. On multivariable analysis, factors associated with operation included age (OR 1.11, 95% CI 1.01-1.18, p<0.05), massive splenic disruption (OR 3.10, 95% CI 1.61-6.19, p<0.001), hollow viscus injury (OR 11.03, 95% CI 3.46-34.28, p<0.001) and transfusion (OR 7.70, 95% CI 4.54-13.16, p<0.001). Management outside a paediatric trauma centre remained significantly associated with operation, whether it be metropolitan adult trauma centre (OR 4.22 95% CI 1.70-10.52, p<0.01), rural trauma centre (OR 3.72 95% CI 1.83-7.83, p<0.001) or metropolitan local hospital (OR 5.23, 95% CI 1.22-18.93 p<0.05). Comparing the 2 eras, the overall operation rate fell, although not significantly, from 12.9% to 8.7% (OR 1.3, 95% CI 0.89-243 p=0.13) CONCLUSION: While Paediatric Surgeons have wholeheartedly adopted non-operative management, away from paediatric centres, children in NSW are still being operated on for BSI unnecessarily. While the factors at play may be complex, further evaluation of the management and movement of injured children within the broad NSW trauma system is required.
Since the 1980's, paediatric surgeons have increasingly managed blunt splenic injury (BSI) in children non-operatively. However, studies in North America have shown higher operation rates in non-paediatric centres and by adult surgeons. This association has not been examined elsewhere. OBJECTIVE: To investigate the management of BSI in New South Wales (NSW) children, to determine the patient and hospital factors related to the odds of operation. Secondarily, to investigate whether the likelihood of operation varied by year. METHODS:Children age 0-16 admitted to a NSW hospital between July 2000 and December 2011 with a diagnosis of BSI were identified in the NSW Admitted Patient Data Collection, and linked to deaths data from Registry of Births Deaths and Marriages, and Bureau of Statistics. The operation rate was calculated and compared between different hospital types. Univariable analysis was used to determine patient and hospital factors associated with operative management. The difference in the odds of operation between the oldest data (July 2000-December 2005) and most recent (January 2006-December 2011) was also examined. Multivariable logistic regression with stepwise elimination was then performed to determine likelihood of operative management according to hospital category and era, adjusting for potential confounders. RESULTS: 955 cases were identified, with 101(10.6%) managed operatively. On multivariable analysis, factors associated with operation included age (OR 1.11, 95% CI 1.01-1.18, p<0.05), massive splenic disruption (OR 3.10, 95% CI 1.61-6.19, p<0.001), hollow viscus injury (OR 11.03, 95% CI 3.46-34.28, p<0.001) and transfusion (OR 7.70, 95% CI 4.54-13.16, p<0.001). Management outside a paediatric trauma centre remained significantly associated with operation, whether it be metropolitan adult trauma centre (OR 4.22 95% CI 1.70-10.52, p<0.01), rural trauma centre (OR 3.72 95% CI 1.83-7.83, p<0.001) or metropolitan local hospital (OR 5.23, 95% CI 1.22-18.93 p<0.05). Comparing the 2 eras, the overall operation rate fell, although not significantly, from 12.9% to 8.7% (OR 1.3, 95% CI 0.89-243 p=0.13) CONCLUSION: While Paediatric Surgeons have wholeheartedly adopted non-operative management, away from paediatric centres, children in NSW are still being operated on for BSI unnecessarily. While the factors at play may be complex, further evaluation of the management and movement of injured children within the broad NSW trauma system is required.
Authors: Michel Teuben; Roy Spijkerman; Henrik Teuber; Roman Pfeifer; Hans-Christoph Pape; William Kramer; Luke Leenen Journal: Patient Saf Surg Date: 2020-01-03