e-ISSN: 2722-3558 Sriwijaya Journal of Surgery [SJS] https://sriwijayasurgery.
The Lung Organ Failure Score (LOFS) as an Early Predictor of Mortality in Blunt Thoracic Trauma: A Preliminary Validation Study in a Southeast Asian Cohort Charita Ulfah Widyawan1*.
Arie Hasiholan Lumban Tobing2.
Theodorus3 1Department of General Surgery.
Dr.
Mohammad Hoesin General Hospital/Faculty of Medicine.
Universitas Sriwijaya.
Palembang.
Indonesia 2Department of Cardiothoracic and Vascular Surgery.
Dr.
Mohammad Hoesin General Hospital.
Palembang.
Indonesia 3Faculty of Medicine.
Universitas Sriwijaya.
Palembang.
Indonesia
ARTICLE
INFO
Keywords:
Blunt thoracic trauma Lung organ failure score Mortality Prognostic score Risk stratification *Corresponding author:
Charita Ulfah Widyawan E-mail address:
charitaulfahw@gmail.
All authors have reviewed and approved the Anal version of the manuscript.
https://doi.
org/10.
37275/sjs.
ABSTRACT
Introduction: Blunt thoracic trauma is a leading cause of trauma-related Early and accurate risk stratification is essential for optimizing outcomes, yet many prognostic scores lack validation in diverse populations.
This study aimed to perform a preliminary evaluation of the prognostic value of the lung organ failure score (LOFS) for predicting in-hospital mortality in patients with blunt thoracic trauma in an Indonesian tertiary trauma center.
Methods: This retrospective, single-center, exploratory cohort study included adult patients (Ou18 year.
admitted with blunt thoracic trauma to Dr.
Mohammad Hoesin General Hospital from January 2023 to January Data on demographics, injury characteristics, initial physiological parameters, and clinical outcomes were collected.
The LOFS was calculated for each patient.
The primary outcome was in-hospital mortality.
Statistical analyses included bivariate comparisons.
Kendall's Tau correlation, and Receiver Operating Characteristic (ROC) curve analysis.
Results: A total of 32 patients were included.
The overall in-hospital mortality rate was 21.
The non-survivor group had a significantly higher mean LOFS than the survivor group .
00A5.
29 vs.
16A3.
92, p=0.
LOFS
demonstrated a moderate, positive correlation with mortality (Kendall's Tau r= 0.
568, p=0.
ROC analysis showed that LOFS had excellent discriminative ability for mortality, with an Area Under the Curve (AUC) of 840 .
% CI: 0.
685Ae0.
995, p=0.
An optimal cut-off score of Ou18 yielded a sensitivity of 85.
7% and a specificity of 80.
Conclusion: In this preliminary study.
LOFS was strongly associated with in-hospital mortality and demonstrated excellent discriminative performance.
The findings suggest LOFS is a promising and simple tool for early risk stratification in this high-risk population.
However, the study's small sample size precluded a reliable assessment of its independence from other risk factors.
Further validation in larger, prospective multicenter studies is essential to confirm these findings.
Introduction Trauma remains a global public health scourge and motor vehicle collisions, falls, and direct impactsAiare responsible for over 70% of these chest injuries.
The the leading cause of death for individuals under the age of 45.
Within this broad spectrum of injury, approaching 60% in severely injured cohorts, a thoracic trauma holds a particularly ominous position, statistic that underscores the profound need for rapid, contributing to approximately one in four trauma- accurate, and decisive clinical management.
related deaths and significantly complicating the The clinical challenge posed by blunt thoracic clinical trajectory of up to 50% of all polytrauma trauma lies in its complex and often insidious 3 The initial mechanical impact, or Blunt force mechanismsAiresulting from "first hit," inflicts direct damage to the chest wall .
uch The lung organ failure score (LOFS), first proposed as rib fractures and flail ches.
and underlying viscera, by Wutzler et al.
, was specifically engineered to fill this most notably the lung parenchyma, leading to It was designed to predict the likelihood of severe pulmonary contusion and laceration.
This primary pulmonary organ failure in polytrauma patients with injury precipitates a cascade of deleterious events:
concomitant chest injuries.
9 Its strength lies in its pain-induced composite nature, integrating ten readily available parameters at admission.
These include demographic Compounding this local insult is the factors .
ge, gende.
, measures of anatomical injury systemic inflammatory response syndrome (SIRS), a severity (ISS.
Abbreviated Injury Scale [AIS] scores for global biological firestorm triggered by tissue damage the thorax and hea.
, and markers of physiological and shock.
4 This systemic inflammation dramatically insult and the iatrogenic burden of resuscitation .
nitial fluid volume, need for emergency and multiple already-injured lungs exquisitely vulnerable to fluid By synthesizing these distinct domains of risk.
LOFS aims to provide a more holistic and respiratory distress syndrome (ARDS), a primary nuanced prediction of lung-related complications than driver of late mortality in trauma patients.
its predecessors.
ventilation-perfusion Consequently, a cornerstone of modern trauma Despite its theoretical appeal, the validation and care is the ability to predict, at the earliest possible application of LOFS have been largely confined to the European populations in which it was developed.
Its Early identification of high-risk utility for predicting the ultimate outcome of mortality, rather than just organ failure, remains under- interventions, such as consideration for early intensive This is particularly true in developing care unit (ICU) admission, aggressive multimodal pain nations and diverse demographic settings, such as control .
uch as regional anesthesi.
, optimized fluid Southeast Asia, where differences in injury patterns, resuscitation strategies, and advanced respiratory healthcare resources, and patient physiology may Over the decades, numerous scoring systems impact the score's performance.
A previous Indonesian have been developed to aid in this prognostic study by Setiawan et al.
showed LOFS to be superior General trauma scores like the injury to TTSS for predicting ventilator use, but its direct severity score (ISS) and the revised trauma score (RTS) prognostic value for mortality has not been specifically provide a global assessment of anatomical and The novelty of this research lies in it being the first specificity to capture the unique pulmonary risks study to specifically validate and quantify the utility of associated with chest trauma.
7 An ISS of 25 from LOFS as a mortality prediction tool in this distinct orthopedic injuries, for instance, carries a vastly patient population, with the goal of providing an different prognosis than an ISS of 25 driven by a severe evidence-based framework for early risk stratification bilateral pulmonary contusion.
Thorax-specific scores, in a resource-constrained healthcare environment.
including the thorax trauma severity score (TTSS).
Therefore, this study aims to conduct a preliminary have been introduced to address this gap, but their evaluation of the prognostic value of the lung organ complexity or inconsistent validation across diverse failure score (LOFS) for predicting in-hospital mortality in patients with blunt thoracic trauma at a tertiary This leaves an unmet need for a simple, trauma center in Indonesia.
reliable, and easily applicable tool for risk stratification in this specific patient population.
Methods .
ale/femal.
, and Body Mass Index (BMI), calculated This study was a retrospective, single-center.
Dr.
as weight .
/ height .
Clinical History: Preexisting .
Mohammad Hoesin General Hospital in Palembang, mellitus, chronic pulmonary diseas.
and smoking South Sumatra.
Indonesia.
As a national referral .
Injury Characteristics: The mechanism of hospital and the primary tertiary care trauma center injury was recorded.
Injury severity was coded using for the province, it manages a high volume of complex the Abbreviated Injury Scale (AIS), 2005 version.
The injury cases.
The study protocol was reviewed by the Injury Severity Score (ISS) was calculated as the sum Institutional Review Board, which granted an ethical of the squares of the highest AIS scores in the three exemption due to the retrospective and de-identified most severely injured body regions.
Severe thoracic injury was defined as an AIS score for the thorax (No.
DP.
03/D.
XVi.
08/ETIK/218/2.
The region of Ou4, and severe head injury was defined as an requirement for individual informed consent was AIS score for the head/neck region of Ou3.
All AIS and This report was prepared in accordance with ISS calculations were performed retrospectively by the Strengthening the Reporting of Observational trained personnel based on a comprehensive review of Studies in Epidemiology (STROBE) guidelines.
all imaging reports (CT.
X-ra.
and operative notes.
The study population included all patients with a Initial Resuscitation and Management: Glasgow Coma diagnosis of blunt thoracic trauma admitted to the Scale (GCS) score on arrival, initial arterial lactate level hospital between January 1st, 2023, and January 31st, .
mol/L), and base deficit .
mol/L).
The total volume Patients were identified using the hospitalAos medical records database and relevant International administered within the first 6 hours of emergency Classification of Diseases, 10th Revision (ICD-.
department arrival was recorded in liters.
The number codes for thoracic injuries.
Inclusion criteria were: .
of packed red blood cell (PRBC) units transfused in the adult patients aged 18 years or older at admission, and first 24 hours was noted.
The performance of any .
a primary or significant secondary diagnosis of emergency surgery, defined as an urgent, non-elective .
rystalloids life-threatening Exclusion criteria were: .
patients with isolated penetrating thoracic trauma .
uch as stab or gunshot wound.
patients who were dead on arrival or died in the emergency department prior to The total number of surgical interventions comprehensive diagnostic evaluation, precluding full during the hospital stay was also collected.
Hospital LOFS calculation.
pregnant patients.
Course and Outcomes: The primary outcome was in- hospital mortality, defined as death from any cause insufficient for LOFS calculation or primary outcome during the index hospitalization.
Secondary outcomes included the need for mechanical ventilation and the .
ncluding A standardized data collection instrument was total length of hospital stay .
n day.
The LOFS was utilized to retrospectively extract data from electronic calculated for each patient according to the original and paper-based medical records.
To ensure data methodology described by Wutzler et al.
Points were integrity, two researchers independently performed assigned for ten independent predictors, and the total the data abstraction, with a senior surgeon resolving score was derived from their sum, as detailed in Table any discrepancies.
The following variables were .
Demographics: Age .
n year.
, gender All data were analyzed using IBM SPSS Statistics was used for non-normally distributed continuous for Windows.
Version 25.
0 (Armonk.
NY: IBM Cor.
The Chi-square test or Fisher's exact test Continuous variables were assessed for normality .
or cell counts <.
was used for categorical variables.
using the Shapiro-Wilk test, appropriate for the The correlation between the LOFS .
s a continuous sample size (<.
Normally distributed data were variabl.
and the dichotomous outcome of mortality presented as mean A standard deviation (SD), while was assessed using the Kendall's Tau correlation non-normally distributed data were presented as This non-parametric test was chosen due median and interquartile range (IQR).
Categorical to the small sample size and non-normally distributed variables were described using frequencies and data, providing a robust measure of association.
To compare characteristics between the evaluate the overall discriminative performance of survivor and non-survivor groups, the independent LOFS samples t-test was used for normally distributed Characteristic (ROC) curve was constructed.
The Area continuous variables, and the Mann-Whitney U test Under the Curve (AUC) with its 95% confidence Receiver Operating interval (CI) was calculated.
An AUC of 0.
7Ae0.
8 is tailed p-value of <0.
05 was considered statistically considered acceptable, 0.
8Ae0.
9 is excellent, and >0.
significant for all analyses.
is outstanding.
The optimal cut-off value for LOFS that maximized the Youden's index (Sensitivity Specificity - .
was determined to provide a clinically relevant Results During the two-year study period, 38 patients were Given the small number of mortality events .
, a multivariate logistic regression analysis was Following the application of exclusion criteria, 6 not performed.
Such an analysis would be statistically patients were excluded: 5 were under the age of 18, invalid due to severe overfitting .
ewer than 10 events and 1 had a significant concomitant penetrating per predictor variabl.
, yielding unreliable estimates of This resulted in a final cohort of 32 patients for independent association.
Therefore, this study focuses The patient selection process is outlined in on associative and discriminative analyses only.
A two- Figure 1.
Figure 1.
STROBE flow diagram of patient selection.
The demographic and clinical characteristics of the high burden of injury, reflected by a median GCS of study cohort are detailed in Table 2.
The mean age was 13, a median ISS of 25.
0, and evidence of significant 00A16.
physiological derangement with a median lactate of 3.
1%).
The cohort presented with a mmol/L.
The most common mechanism of injury was motor vehicle collision .
6%).
A high percentage of emergency surgery in 75.
0% of cases and mechanical ventilation in 46.
9% of the cohort.
The overall in- .
0%),
hypertension was the most common comorbidity .
0%).
The hospital mortality rate was 21.
9% .
of 32 patient.
Table 3 Anatomically, the injury burden was substantially characteristics between patients who survived and higher among non-survivors, with a significantly those who died.
The non-survivor group demonstrated higher median ISS .
0 vs.
0, p=0.
and a evidence of more profound shock and severe injury on markedly higher prevalence of severe thoracic injury They had a significantly lower median GCS .
(AIS Thorax Ou 4: 85.
7% vs.
0%, p=0.
14, p=0.
, higher median lactate .
8 vs.
Consequently, the need for mechanical ventilation was mmol/L, p=0.
, and required a greater volume of significantly greater in the non-survivor group .
initial fluid resuscitation .
00 L vs.
00 L,
0%, p=0.
While non-survivors tended to be p=0.
older, this difference was not statistically significant.
The primary analysis revealed a profound and was further quantified using correlation analysis.
The highly statistically significant difference in LOFS Kendall's Tau test revealed a moderate, positive, and between the two outcome groups.
The mean LOFS for significant correlation between the LOFS and mortality non-survivors was 21.
00 A 5.
29, compared to 14.
= 0.
568, p=0.
This indicates that as the LOFS increases, the risk of death increases proportionally.
demonstrates that higher LOFS values, calculated The overall ability of LOFS to discriminate between from admission data, are strongly associated with survivors and non-survivors was evaluated using ROC subsequent in-hospital mortality.
This relationship curve analysis (Figure .
The LOFS demonstrated .
<0.
This excellent predictive accuracy, with an Area Under the predictive value (NPV) of 95.
The high NPV is Curve (AUC) of 0.
% CI: 0.
685Ae0.
p=0.
particularly noteworthy, suggesting that patients with Based on Youden's index, the optimal cut-off value for a LOFS below 18 are at a very low risk of in-hospital LOFS in predicting mortality in this cohort was The Area Under the Curve (AUC) was 0.
determined to be Ou18.
This threshold yielded a indicating excellent discrimination between survivors sensitivity of 85.
7%, a specificity of 80.
0%, a positive and non-survivors.
predictive value (PPV) of 54.
5%, and a negative Figure 2.
Receiver operating characteristic (ROC) curve for LOFS in predicting in-hospital mortality.
Discussion tool for mortality in blunt thoracic trauma patients This study represents the first focused evaluation within an Indonesian and broader Southeast Asian of the lung organ failure score (LOFS) as a prognostic 11 The principal finding of this preliminary research is that a higher LOFS, calculated from readily through a massive sympathetic surge, causing a rapid available admission data, is strongly associated with and damaging shift of fluid into the pulmonary in-hospital interstitium and alveoli.
15 Furthermore, the injured discriminative capability.
Patients who did not survive brain releases a torrent of pro-inflammatory cytokines had a mean LOFS of 21, a full 7 points higher than the mean score of 14 in those who survived.
The AUC of propagating a systemic inflammatory state that 840 indicates that the score is highly effective at synergistically worsens the primary lung injury and distinguishing between patients who will live and heightens ARDS risk.
This destructive crosstalk those who will die.
These findings suggest that LOFS explains why concomitant TBI is such a potent holds considerable promise as a simple, early, and harbinger of poor outcomes, a principle effectively effective risk stratification tool in this high-risk captured by the LOFS.
The robust performance of LOFS can be LOFS blood-brain attributed to its sophisticated integration of multiple, physiological insult and the subsequent therapeutic distinct pathophysiological domains that converge to The volume of initial fluid resuscitation is a determine a patientAos It moves beyond a criticalAiand controversialAicomponent simple anatomical summary of injuries to create a trauma care.
While essential for treating hemorrhagic more holistic portrait of the patient's condition.
shock, aggressive fluid administration is a double- The score heavily weights anatomical injury edged sword.
Our study confirmed that non-survivors severity, specifically through the ISS and AIS scores received significantly more fluid in the first six hours.
for the thorax and head.
13 Our data strongly support Excessive crystalloid infusion can lead to dilutional this, showing that severe thoracic injury was over four times more prevalent in non-survivors .
7% vs.
0%).
Severe blunt thoracic trauma represents the This iatrogenic "resuscitation injury" quintessential "first hit": direct mechanical violence to exacerbates capillary leak, leading to generalized the lung parenchyma and chest wall that initiates a tissue edema.
In the lungs, this manifests as vicious cycle of pain, hypovilation, atelectasis, and worsening pulmonary edema, which impairs gas localized inflammation.
This localized damage primes the lung for further injury.
The overall trauma burden, contributes to respiratory failure and death.
quantified by the ISS, triggers a systemic inflammatory Finally, interventionAi response (SIRS), which constitutes the initial phase of especially emergency surgery or multiple proceduresAi the "second hit".
This systemic inflammation enhances is a powerful surrogate for both injury severity and the capillary permeability globally, but its effects are most physiological stress of the ongoing "second hit".
devastating in the already-compromised pulmonary major operation, though life-saving, induces its own vasculature, paving the way for ARDS and multiple significant inflammatory response, adding fuel to the organ dysfunction syndrome (MODS), the leading post-traumatic SIRS cascade.
This can be the final causes of late trauma mortality.
insult that tips a patient with compromised pulmonary The inclusion of severe head injury (AIS Head Ou .
reserve into overt, irreversible ARDS.
The inclusion of in LOFS is a particularly insightful component, age and male gender reflects established demographic acknowledging the critical importance of the brain- risk factors, with advancing age corresponding to lung axis.
A severe traumatic brain injury (TBI) can (NPE) Figure 3.
Pathophysiological rationale and integration of risk domains.
Our findings, which are remarkably consistent with the original developmental study by Wutzler et al.
concerning the predictive factors for pulmonary mortality, as its effect cannot be reliably disentangled from other confounding factors.
Second, the exclusion of patients who died in the failure, extend their work by demonstrating that this same constellation of factors, synthesized by the evaluation introduces a potential for survivorship bias.
LOFS, is also highly predictive of mortality.
18 The These patients represent the most severe spectrum of identified optimal LOFS cut-off of Ou18 in our cohort injury, and their exclusion means our findings are only provides a practical, albeit preliminary, threshold for applicable to patients who survive the initial phase of A patient presenting with a score above this This may lead to an underestimation of value warrants heightened vigilance and could be a the overall mortality associated with blunt thoracic trigger for a pre-defined care pathway.
This might trauma and could potentially inflate the score's performance metrics.
Third, the findings from our ICU pulmonary toilet, protocolized pain management with immediately generalizable to other populations or regional anesthetic techniques, and a judicious fluid healthcare systems with different injury patterns, pre- management strategy.
The scoreAos high negative predictive value .
2%) is perhaps its most powerful Finally, the optimal LOFS cut-off value of a score below 18 can provide a degree of Ou18 identified in this study should be considered exploratory and requires rigorous validation in a allocation of less resource-intensive care, a critical larger, more diverse cohort before it can be confidently consideration in any healthcare environment.
recommended for widespread clinical implementation.
systems, or consideration for invasive monitoring, aggressive Indonesian in-hospital The future of trauma prognostication lies in moving from population-based estimates to individualized risk While LOFS is a step in this direction, the large-scale.
Conclusion In this preliminary retrospective study, the Lung Organ Failure Score in-hospital study would be necessary to confirm our findings, mortality in adult patients with blunt thoracic trauma.
refine the optimal cut-off value, and definitively A higher LOFS, calculated from simple admission establish the score's independent predictive value parameters, was strongly associated with an increased through a robust multivariate analysis.
Furthermore, risk of death.
These findings suggest LOFS is a future research could explore the integration of novel promising and simple tool for early risk stratification, biomarkers of inflammation .
uch as interleukins and capable of integrating key anatomical, physiological, cell-free DNA) or endothelial injury into the LOFS and therapeutic variables into a single, clinically framework to potentially enhance its predictive power.
useful metric.
Its implementation may help clinicians This study has several important limitations that prospective, multicenter validation study.
20 Such a high-risk must be acknowledged.
The primary limitation is its monitoring and targeted interventions.
However, due preliminary, single-center, retrospective design and, to the study's small sample size, its independent most significantly, its small sample size.
The cohort of predictive value could not be determined.
Further 32 patients with only 7 mortality events provided validation in larger, prospective, multicenter studies is insufficient statistical power to perform a multivariate essential to confirm these findings and establish its Therefore, while our study demonstrates a definitive role in clinical practice.
strong association and excellent discrimination, it cannot establish LOFS as an independent predictor of References