Performance and outcome of ressucitative thoracotomies in a southern Brazil trauma center: a 7-year retrospective analysis

ABSTRACT Objective: the study aims to analyze the performance and outcome of resuscitation thoracotomy (TR) performed in patients victims of penetrating and blunt trauma in a trauma center in southern Brazil during a 7 years period. Methods: retrospective study based on the analysis of medical records of patients undergoing TR, from 2014 to 2020, in the emergency service of the Hospital do Trabalhador, Curitiba - Paraná, Brazil. Results: a total of 46 TR were performed during the study period, of which 89.1% were male. The mean age of patients undergoing TR was 34.1±12.94 years (range 16 and 69 years). Penetrating trauma corresponded to the majority of indications with 80.4%, of these 86.5% victims of gunshot wounds and 13.5% victims of knife wounds. On the other hand, only 19.6% undergoing TR were victims of blunt trauma. Regarding the outcome variables, 84.78% of the patients had declared deaths during the procedure, considered non-responders. 15.22% of patients survived after the procedure. 4.35% of patients undergoing TR were discharged from the hospital, 50% of which were victims of blunt trauma. Conclusion: the data obtained in our study are in accordance with the world literature, reinforcing the need for a continuous effort to perform TR, respecting its indications and limitations in patients victims of severe penetrating or blunt trauma.

Survival rates reported after RT vary widely, which may be due to different characteristics between pre-hospital services, population distribution, mechanism of injury, and sample size 8 .
Thus, the present study aims to retrospectively analyze the performance and outcome of RT in victims of penetrating and blunt trauma in a trauma referral hospital in Southern Brazil, in the last 7 years.

Study design and period
We conducted a retrospective, observational, and descriptive study that analyzed the medical records of patients undergoing RT, from 2014 to 2020, in the emergency service of the Hospital do Trabalhador, Curitiba -Paraná state, Brazil.

Definitions
We defined RT as an immediate thoracotomy performed on an emergency basis in the emergency room or in the operating room in patients with absence of pulse after blunt or penetrating trauma, with or without signs of life 1 .The following parameters were considered as signs of life: pupillary response, spontaneous breathing, presence of carotid pulse, measurable or palpable blood pressure, extremity movement, and cardiac electrical activity 9 .Thus, we included patients who met the proposed definition, and excluded those with insufficient data in their medical records.
The Injury Severity Score (ISS) is an anatomical trauma score that considers the affected anatomical region and the degree of severity in each region, ranging from 1 to 5. Furthermore, the most severe injury score of the three most injured segments is squared, with trauma being classified as mild (ISS <9), moderate (ISS 9-14), and severe (ISS >14), ranging from 1 to 75 10 .In cases where a segment is classified as an untreatable lesion (6), the score automatically assumes a value of 75 10 .
The Revised Trauma Score (RTS) uses the Glasgow Coma Scale (GCS), respiratory rate, and systolic blood pressure, ranging from 0 to 12.A lower score predicts lower survival 11 .
The Glasgow Coma Scale (GCS) is used to identify a patient's level of consciousness, assessing the capacity for motor response, verbal response, and eye opening, ranging from 3 to 15 points 12 .
The Trauma Injury Severity Score (TRISS) is a method used in the retrospective analysis of the survival probability of trauma patients, which uses a mathematical equation including RTS, ISS, and patient age 13 .In addition, it allows to determine the quality of service provided at the institution and to compare it with that of other trauma centers 13 .

Variables analyzed
We assessed sex, trauma mechanism, clinical data at admission, outcome, postoperative complications, mortality, and hospital discharge.For the classification of trauma severity, we used ISS, TRISS, RTS, and GCS.

Statistical analysis
We analyzed quantitative variables with the Mann-Whitney test, and qualitative ones, with the Fisher's exact test.We performed all statistical tests with the GRAPHPAD PRISM ® statistical package, considering a significance level of 5% (p<0.05).

Research Ethics
The Ethics in Research Committee approved the study with the opinion 4.369.508, on October 29 th , 2020.Ethics Assessment Presentation Certificate (CAAE) 39591220.0.0000.5225.
There was a predominance of the young population, with 69.5% of patients aged ≤40 years.Penetrating trauma was responsible for most RT indications (80.4%), of which 86.5% were due to gunshot wounds (GSW), and 13.5%, stab wounds (SW).On the other hand, 19.6% of individuals undergoing RT were victims of blunt trauma (Table 1).Regarding outcome, 84.8% of patients died during the procedure, and were considered nonresponders.On the other hand, RT responders (those who responded to initial resuscitation measures and came out alive after the operation) had a survival rate after the procedure of 15.2%.However, 71.4% of responding patients died (Table 2).Therefore, we obtained a hospital discharge rate of 4.4% (Figure 2).

DISCUSSION
In the last 40 years, indications for RT have been improved due to bettered information on results and outcomes after this procedure 14 .Thus, although there is great heterogeneity of indications and applications evidenced by the wide variability in survival rates reported in the medical literature, currently RT has been associated with excellent results in reducing mortality in specific circumstances = .
some authors point out the restricted resources in the emergency room, the costs inherent to the procedure, and the risk of exposure to blood-borne pathogens, such as HIV 16,17 .However, the study by Baker, Thomas, and Trunkey 18 showed that the total benefits of RT were 2.4 times greater than the total cost of the procedure, taking into account economic aspects such as shorter hospital stay and low cost when compared with other procedures.
In parallel, the study by Branney, Moore, and Feldhaus 19 also demonstrated a favorable cost-benefit ratio for RT.
In addition, a study published in 2018 concluded that the risk of exposure of the staff involved in RT is low and that it should not influence the decision to perform the procedure 20 .
In accordance with the literature reported survival rates after RT in blunt trauma of 1.4% and 5.2%, respectively 7,21 .In our study, we had a hospital discharge rate of 2.1% after RT in blunt trauma.For some authors, the mortality rate associated with RT is probably due to the patient's critical condition, as well as to the indication for the procedure 22 .
In line with the literature, our results on ISS, RTS, GCS, and TRISS were more favorable in patients analyzed did not use GCS or another objective scale for the neurological assessment of patients undergoing RT.
In that review, only a subjective qualitative assessment to characterize the patient's neurological status was reported.Thus, in the right patient population, RT cannot only save lives, but also potentially lead to good recovery and function after otherwise fatal injuries 27 .
The unicentric analysis, relatively small sample, retrospective evaluation, and absence of previously defined sample size were limitations of our study.Thus, larger-sample, multicenter, and prospective studies are needed for further elucidation of RT.

CONCLUSION
The  24 reported a mean ISS of 25, a value similar to that found in our analysis (25.85).
Higher RTS values indicate patients with better clinical conditions 11 .In our study, patients who were discharged from the hospital had a mean RTS of 7.70.On the other hand, patients who died had a mean of 2.63.
These results agree with those of Thorsen et al. 8 , who found a mean RTS value for the group that died of 2, and for the group that survived, of 8.
As for the Glasgow Coma Scale, according to Lustenberger et al. 25 a GCS greater than 8 is an independent predictor of survival after RT.In agreement, we observed a mean GCS in patients who died of 5.45, and for those who were discharged from the hospita l,14 .
Similarly, Thorsen et al. 8 observed that patients who died and those who survived after RT had mean GCS values of 2 and 12, respectively.However, the systematic review by Narvestad et al. 26

R E S U M O R E S U M O
ISS: injury severity score.TRISS: trauma injury severity score.RTS: revised trauma score.GCS: glasgow coma scale.SBP: systolic blood pressure.*p<0.05.

Table 1 .
Epidemiology and clinical characteristics of deaths and hospital discharges after resuscitative thoracotomy.

Table 2 .
Clinical characteristics of responders to resuscitative thoracotomy..
1SS: injury severity score.TRISS: trauma injury severity score.RTS: revised trauma score.GCS: glasgow coma scale.SBP: systolic blood pressure.orthosewithsigns of life present after blunt injury.Also conditionally, RT is not recommended for patients with absence of pulse and vital signs after blunt injury1.Furthermore, as limiting factors for performing RT, Rev Col Bras Cir 49:e20223146 Haida Performance and outcome of ressucitative thoracotomies in a southern Brazil trauma center: a 7-year retrospective analysis analysis of RT performed at a trauma referral center in southern Brazil points to results similar to those reported in the world literature.Data from our study support the need for continuous effort to perform RT, respecting the indications and limitations in victims of severe penetrating or blunt trauma.who survived.Onat et al. 2 had a mean of 30.71 points in patients who underwent RT and died.This result is similar to what we found (30.27 points).Regarding patients who responded to RT, Ito et al.