
Multiple organ dysfunction after trauma
Author(s) -
Cole E.,
Gillespie S.,
Vulliamy P.,
Brohi K.
Publication year - 2020
Publication title -
british journal of surgery
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.202
H-Index - 201
eISSN - 1365-2168
pISSN - 0007-1323
DOI - 10.1002/bjs.11361
Subject(s) - medicine , multiple organ dysfunction syndrome , resuscitation , injury severity score , mortality rate , sofa score , population , cluster (spacecraft) , organ dysfunction , surgery , emergency medicine , poison control , injury prevention , sepsis , environmental health , computer science , programming language
Background The nature of multiple organ dysfunction syndrome (MODS) after traumatic injury is evolving as resuscitation practices advance and more patients survive their injuries to reach critical care. The aim of this study was to characterize contemporary MODS subtypes in trauma critical care at a population level. Methods Adult patients admitted to major trauma centre critical care units were enrolled in this 4‐week point‐prevalence study. MODS was defined by a daily total Sequential Organ Failure Assessment (SOFA) score of more than 5. Hierarchical clustering of SOFA scores over time was used to identify MODS subtypes. Results Some 440 patients were enrolled, of whom 245 (55·7 per cent) developed MODS. MODS carried a high mortality rate (22·0 per cent versus 0·5 per cent in those without MODS; P < 0·001) and 24·0 per cent of deaths occurred within the first 48 h after injury. Three patterns of MODS were identified, all present on admission. Cluster 1 MODS resolved early with a median time to recovery of 4 days and a mortality rate of 14·4 per cent. Cluster 2 had a delayed recovery (median 13 days) and a mortality rate of 35 per cent. Cluster 3 had a prolonged recovery (median 25 days) and high associated mortality rate of 46 per cent. Multivariable analysis revealed distinct clinical associations for each form of MODS; 24‐hour crystalloid administration was associated strongly with cluster 1 ( P = 0·009), traumatic brain injury with cluster 2 ( P = 0·002) and admission shock severity with cluster 3 ( P = 0·003). Conclusion Contemporary MODS has at least three distinct types based on patterns of severity and recovery. Further characterization of MODS subtypes and their underlying pathophysiology may lead to future opportunities for early stratification and targeted interventions.