Predicting the Outcome of Chest Injury Patients by Thorax Trauma Severity Score: An Observational Study

INTRODUCTION

Among the various causes of trauma-related mortality, thoracic trauma ranks as the third leading cause of mortality following head and spinal injury.¹ Multiple trauma scoring systems have been developed to determine the injury severity, prognosticate patient outcomes, and predict associated mortality and morbidity.² Initially, the advanced trauma life support (ATLS) protocols are used for the assessment, stabilization, and management of all trauma patients, focussing on sequential evaluation and stabilization of airway, breathing, circulation, and disability. Basic radiological investigations, such as chest X-ray, pelvic X-ray, and extended focussed assessment with sonography in trauma (e-FAST), are also part of the initial assessment to facilitate prompt detection of injuries.³

The initial scoring systems, such as the trauma and injury severity score (TRISS), injury severity score, and revised trauma score, were designed to assess trauma comprehensively;⁴ however, they were not specifically focused on chest trauma. Although TRISS is used most commonly to predict the outcome in a trauma patient, it is not specific to chest trauma patients. Thus, it underestimates the severity of chest trauma.⁵ To overcome these shortcomings, Pape et al. developed the thorax trauma scoring system (TTSS), which included both the anatomical and functional parameters of thoracic trauma.⁶ TTSS includes parameters such as arterial blood gas (ABG), number of rib fractures, lung contusion, pleural involvement, and age, with a scoring range from 0 to 25 [Table 1].

Together, all these parameters give a comprehensive basis for quantifying the severity of rib fracture complications. At present, there is limited information on the proportion of patients in the western region of India who develop respiratory complications following chest trauma and the morbidity and mortality associated with it. Therefore, this study aims to use this validated and well-established scoring system to predict morbidity and mortality in chest trauma patients at our hospital.

MATERIALS AND METHODS

This hospital-based prospective observational study was conducted in the Department of General Surgery at a tertiary care teaching institute in India from January 2020 to December 2021. Patients presenting to the emergency department with blunt or penetrating chest injuries, including hemothorax, pneumothorax, hemopneumothorax, and thoracic visceral injuries with or without rib fractures, were included in the study. Patients with rib fractures due to iatrogenic causes or those under 18 years of age were excluded from the study. Patients who had severe abdominal injuries requiring surgery were also excluded from the study. A total of 100 patients were enrolled during this period. All patients received in the emergency department were managed according to the standard ATLS protocol. It was followed by X-ray, e-FAST, and computed tomography whenever indicated. Additional hematological and biochemical investigations, along with blood grouping, cross-matching, and ABG were also done. Treatment of chest injury, whether conservative or invasive (i.e., tube thoracostomy and thoracotomy), was performed as indicated, supplemented with oxygen therapy, chest physiotherapy, and mechanical ventilation if required. The TTSS was calculated for each patient at admission. Patients were grouped as per a TTSS score range. Group I includes patients with a TTSS score of <7 and Group II patients with a TTSS score of >7. The association between TTSS score and outcomes was studied. Pain severity was assessed using the numerical pain rating score (NPRS) on admission, day 1, day 3, day 7, at discharge, and 1 month after discharge. The NPRS is a pain assessment scale where 0 indicates no pain and 10 represents the worst pain imaginable. Analgesics (oral/intravenous/intramuscular), intercostal nerve block, and epidural analgesia were used for pain management. Any additional complications or systemic injuries during the hospital stay were also documented.

RESULTS

A total of 100 patients were recruited during the study period, comprising 85 males and 15 females, with a mean age of 38 ± 14 years. All patients sustained thoracic injuries, often accompanied by other multi-visceral injuries. Thirty patients required surgery for either thoracic or extra-thoracic reasons, predominantly orthopedic. Out of the 100 chest trauma patients, 11 died. The primary mode of injury was RTAs, accounting for 68%, followed by fall from height (16%), blunt trauma (7%), bull horn injuries (5%), penetrating trauma (2%), and assault (2%).

Admission severity and events

Of the five parameters of TTSS, 78% of the 100 patients had unilateral rib fractures, and 22% had bilateral rib fractures. Hemothorax was present in 68 patients, out of which unilateral hemothorax was present in 55 patients, while 13 patients had bilateral hemothorax. Pneumothorax was present in 69 patients, unilateral pneumothorax in 57 patients, and bilateral pneumothorax was present in 12 patients. Tension pneumothorax was there in five patients. A total of 58 patients had a pulmonary contusion; 46 had unilateral lung contusion, and 12 had bilateral lung contusion. The flail segment was present in 17 patients. Table 2 summarizes the chest wall injuries.

Table 2: Description of chest wall injuries.

Chest wall injuries	No. of patients
Rib fractures	100
Unilateral	78
Bilateral	22
Haemothorax	68
Unilateral	55
Bilateral	13
Pneumothorax	69
Unilateral	57
Bilateral	12
Tension pneumothorax	5
Lung contusion	58
Unilateral	46
Bilateral	12
Flail segment	17

Most of the patients in the study required tube thoracostomy only (non-operative management) for the treatment of chest trauma. Furthermore, after surgical intervention, the chest tube was kept for monitoring of these patients. A total of 82 patients needed tube thoracostomy placement after chest trauma, out of which 66 patients needed unilateral tube thoracostomy insertion, and the rest 16 patients needed bilateral chest tube insertion. The most common indication of chest tube insertion was unilateral hemopneumothorax, followed by unilateral pneumothorax. The other indications of chest tube insertion were unilateral hemothorax, bilateral hemopneumothorax, bilateral hemothorax and bilateral pneumothorax. Eighteen patients did not require placement of a chest tube. These patients had mostly rib fractures with or without minimal hemothorax or pneumothorax, which was managed with analgesics, either in the form of oral, intravenous, or in the form of epidural analgesics.

The TTSS ranged from a minimum of 1 to a maximum of 16, with the highest number of patients (49 in number) having TTSS values between 6 and 10 [Figure 1].

Close

Figure 1:

Bar chart showing the thorax trauma severity score (TISS) in patients at admission.

Export to PPT

Among patients with a TTSS of 7 or lower, the median HLOS was 6 days (interquartile range [IQR]: 4–8 days); and eight patients required ICU care, with a median ICU LOS of 5 days (IQR: 3–10.75 days). For patients with a TTSS >7, the median HLOS was 7 days (IQR: 3.5–11 days), and 12 patients required ICU care, with a median ICU LOS of 6.5 days (IQR: 2.5– 13 days). The outcomes of patients in comparison to TTTS are depicted in Table 3. The P-value for the association between TTSS and HLOS was 0.150, and for TTSS and ICU LOS, it was 0.739, both statistically non-significant. Sixty-three patients returned to normal work within 30 days of discharge. The Chi-square test yielded P < 0.001 (confidence interval [CI] = 95%), indicating statistical significance. Table 4 summarizes the outcomes of patients.

Table 3: Comparison of pneumonia and mechanical ventilation with TTSS.

Pneumonia patient	Median TTSS (IQR)
Patients who developed pneumonia	12.00 (10.34–12.77)
Patients who did not develop pneumonia	7.00 (5.93–7.09)
Patients who needed mechanical ventilation	10.00 (7.49–12.04)
Patients who did not need mechanical ventilation	7.00 (6.80–8.17)

IQR: Interquartile range, TTSS: orax trauma severity score

Table 4: Outcomes of patients.

Hospital stay in days (median) (IQR)
TTSS ≤7	6 (4–8)
TTSS >7	7 (3.5–11)
Patients requiring ICU stay	20
Median stay in ICU with IQR
TTSS ≤7	5 (3–10.75)
TTSS >7	6.5 (2.5–13)
Patients who underwent surgery	30
Patients developing pneumonia	27
Need for mechanical ventilation	17
Normal return to activity within 30 days	63
Death	11

IQR: Interquartile range, ICU: Intensive care unit, TTSS: Thorax traumaseverity score

Patients who developed pneumonia during their hospital stay had higher TTSS values [Table 3]. A total of 27 patients developed pneumonia, with a median TTSS of 12 (IQR: 10.34–12.77). Among patients without pneumonia, the median TTSS was 7 (IQR: 5.93–7.09), with P < 0.001. Seventeen patients required mechanical ventilation. The median TTSS for those needing mechanical ventilation was 10 (IQR: 7.49–12.04), while the median for those not requiring it was 7 (IQR: 6.80–8.17), with P = 0.009 (CI = 95%).

The overall mortality among chest trauma patients was 11%. The ROC was used to calculate the sensitivity and specificity of the TTSS to predict hospital mortality. The area under the curve [Figure 2] shows a value of 0.83. The patients who died of thoracic injuries had higher TTSS scores. With a TTSS cutoff of 9.5, it can predict mortality with a sensitivity of 90.9% and specificity of 78.7%. Youden’s index for this TTSS value was 69.6%.

Close

Figure 2:

Receiver operating characteristic (ROC) curve for thorax trauma severity score and mortality.

Export to PPT

DISCUSSION

In the year 2000, Pape et al. first introduced the TTSS specifically for chest injuries. It incorporates both anatomical and functional aspects aimed at assisting emergency medical treatment and evaluation by identifying chest injury patients at risk of pulmonary complications.⁶ The TTSS has since been validated for predicting HLOS, ICU LOS, and mortality.^7-9

In our study, blunt thoracic trauma was the primary cause of injury, followed by fall from height. RTAs were the major mechanism of blunt trauma. This aligns with numerous previous studies where RTI was identified as the leading cause of injury.^1,9 When comparing TTSS with HLOS, no significant correlation was found, likely due to other systemic injuries in polytrauma patients that required prolonged hospitalization, particularly orthopedic injuries necessitating bone fixation. Many patients also sustained neurological injuries, which extended the hospital or ICU stays despite relatively minor thoracic injuries. Similarly, ICU LOS was not correlated with TTSS, as many patients required extended ICU care due to neurosurgical injuries (particularly head trauma), even if their chest injuries were minor. Some patients also had diffuse axonal injury, further prolonging ICU care. This finding contrasts with results published by Bagaria et al., who reported a significant association between higher TTSS scores and prolonged ICU stay.⁸ The probable reason for this difference may be because only the chest trauma patients who were mechanically ventilated were included by the authors in their study.

While many patients had multiple systemic injuries, some sustained isolated thoracic injuries. Patients were followed up for 30 days post-discharge to observe their ability to resume normal daily activities. Sixty-three patients returned to their usual activities, while most of the remaining patients had orthopedic injuries. Those unable to resume daily activities had either undergone spine fixation surgery or fracture reduction and were advised limited mobility. A lower TTSS was associated with an earlier return to normal activities.

TTSS was significantly associated with the development of pneumonia in chest trauma patients, with a total of 27 patients developing pneumonia. For these patients, the median TTSS was 12, aligning with findings by Aukema et al., who reported significantly higher TTSS scores in patients who developed acute respiratory distress syndrome (P = 0.005, CI = 95%).⁷ It was noted that older patients were more prone to pneumonia following chest injury.

The TTSS was also effective in predicting the need for mechanical ventilation during the hospital stay. Seventeen patients required mechanical ventilator support, either due to chest injury or for surgical needs. The Chi-square test indicated that TTSS was a statistically significant predictor of mechanical ventilation needs (P = 0.009, CI = 95%). This result aligns with findings by Zahran et al., who noted that a TTSS above 7 was associated with a higher likelihood and duration of mechanical ventilation.¹⁰

Mortality occurred in 11 patients. The area under the ROC curve (AUC) for TTSS in predicting mortality was 0.831, which is comparable to Aukema et al.’s findings, where the AUC was 0.844,⁷ and Kanake et al.’s findings, where the AUC was 0.98.⁹ A TTSS cutoff of 9.5 points yielded a sensitivity of 90.9% and a specificity of 78.7%, suggesting that a TTSS cutoff of 9.5 can effectively predict mortality. This result is somewhat higher than previous findings, where a TTSS of 8 or more demonstrated 92.3% sensitivity and 100% specificity for poor prognosis.⁹ Another study suggested a TTSS cutoff of 8 with a sensitivity of 80% and specificity of 94%.¹¹ Kanake et al. suggested that a TTSS value of 7.5 points or above had 96.9% sensitivity and 92.1% specificity to morbidity and mortality of patients with isolated thoracic trauma after exclusion of associated severe extra thoracic injuries.⁹

Thus, this study guides future research. Further studies are necessary to explore improved scoring systems that can accurately assess the severity of thoracic trauma in patients with significant systemic injuries.

CONCLUSION

The TTSS showed a strong correlation with complications such as pneumonia, the need for mechanical ventilation, and mortality. However, TTSS did not effectively predict the duration of HLOS or ICU LOS. We also conclude that a TTSS cutoff of 9.5 can more accurately predict mortality.