Article Text

Download PDFPDF
Initial focused assessment with sonography in trauma versus initial CT for patients with haemodynamically stable torso trauma
  1. Yutaka Kondo1,2,
  2. Hiroyuki Ohbe1,
  3. Hideo Yasunaga1,
  4. Hiroshi Tanaka2
  1. 1 Department of Clinical Epidemiology and Health Economics, School of Public Health, The University of Tokyo, Tokyo, Japan
  2. 2 Department of Emergency and Critical Care Medicine, Juntendo University Urayasu Hospital, Urayasu, Japan
  1. Correspondence to Yutaka Kondo, Department of Clinical Epidemiology and Health Economics, School of Public Health, The University of Tokyo, Tokyo 1130033, Japan; kondokondou2000{at}yahoo.co.jp

Abstract

Objective Focused assessment with sonography in trauma (FAST) examination is a widely known initial evaluation for patients with trauma. However, it remains unclear whether FAST contributes to patient survival in patients with haemodynamically stable trauma. In this study, we compared in-hospital mortality and length of stay between patients undergoing initial FAST vs initial CT for haemodynamically stable torso trauma.

Methods This was a retrospective cohort study using data from 264 major emergency hospitals in the Japan Trauma Data Bank between 2004 and 2016. Patients were included if they had torso trauma with a chest or abdomen abbreviated injury scale score of ≥3 and systolic blood pressure of ≥100 mm Hg at hospital arrival. Eligible patients were divided into those who underwent initial FAST and those who underwent initial CT. Multivariable logistic regression analysis for in-hospital mortality and multivariable linear regression for length of stay were performed to compare the initial FAST and initial CT groups with adjustment for patient backgrounds while also adjusting for within-hospital clustering using a generalised estimating equation.

Results There were 9942 patients; 8558 underwent initial FAST and 1384 underwent initial CT. Multivariable logistic regression showed no significant difference in in-hospital mortality between the initial FAST and initial CT groups (OR 1.37, 95% CI 0.94 to 1.99, p=0.10). Multivariable linear regression revealed that the initial FAST group had a significantly longer length of stay than the initial CT group (difference: 3.5 days; 95% CI 1.0 to 5.9, p<0.01).

Conclusions In-hospital mortality was not significantly different between the initial FAST and initial CT groups for patients with haemodynamically stable torso trauma. Initial CT should be considered in patients with haemodynamically stable torso trauma.

  • trauma
  • ultrasonography
  • tomography
  • emergency departments

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

What is already known on this subject

  • Focused assessment with sonography in trauma (FAST) has been used for more than 30 years as a primary tool for trauma care in emergency departments. The clinical benefits of FAST have been proven for haemodynamically unstable patients who cannot undergo CT. However, the usefulness of FAST in haemodynamically stable patients with trauma is controversial.

What this study adds

  • In this retrospective cohort study using data from 264 hospitals in the Japan Trauma Data Bank, we found no differences between the initial FAST and initial CT groups regarding mortality of patients with haemodynamically stable torso trauma. Initial CT should be considered in patients with haemodynamically stable torso trauma.

Introduction

Focused assessment with sonography in trauma (FAST) examination is a widely known initial evaluation for patients with trauma. Application of FAST has been taught in the Advanced Trauma Life Support programme since 1978 and is the recommended screening tool of choice for early diagnostic investigations in patients with suspected blunt torso trauma.1 2 In particular, the clinical benefits of FAST have been proven for haemodynamically unstable patients who cannot undergo CT; these patients are likely to undergo emergent surgery or transcatheter arterial embolisation immediately after FAST.3 FAST has been used for more than 30 years as a primary tool for trauma care in emergency departments.4

However, the usefulness of FAST in patients with haemodynamically stable trauma is controversial. A previous retrospective cohort study suggested that positive FAST results could provide valuable prognostic information and that FAST should not be omitted for patients with haemodynamically stable trauma.5 Another study reported that negative FAST results in patients with haemodynamically stable trauma without confirmation by CT might result in missed intra-abdominal injuries because of the low sensitivity of FAST.6

It remains unclear whether FAST contributes to patient survival in patients with haemodynamically stable trauma. No randomised trials have been performed to compare outcomes between initial FAST and initial CT.7 8 The aim of the present study was to clarify whether initial FAST can reduce in-hospital mortality or length of stay compared with initial CT in patients with haemodynamically stable torso trauma.

Materials and methods

Because of the anonymous nature of the data, the requirement for informed consent was waived.

Study design and data collection

This study was a multicentre retrospective cohort study of patients treated between 2004 and 2016 using data from the Japan Trauma Data Bank, which was established in 2003 by the Japanese Association for the Surgery of Trauma (Trauma Registry Committee) and the Japanese Association for Acute Medicine (Committee for Clinical Care Evaluation) as the main parties.9 The database includes nationwide data on patients with trauma, including patient characteristics, vital signs during the prehospital phase and at hospital arrival, examinations and treatments performed, diagnostic codes using the abbreviated injury scale score,10 several severity trauma scores and discharge status.11 Data were collected at 264 major emergency hospitals in Japan that chose voluntarily to participate, and approximately 71% of tertiary-level emergency hospitals in Japan participated in the database.12 Approximately 40% of participating hospitals have ability equal to that of level I trauma centres in the USA.13

Study participants

We included patients with torso trauma with a chest or abdominal abbreviated injury scale score of ≥3 and systolic blood pressure of ≥100 mm Hg at hospital arrival. We excluded those who (1) were younger than 18 years of age, (2) were transferred by helicopter, (3) were transferred by a physician-manned ambulance, (4) were transferred from other hospitals, (5) were discharged to other hospitals on the arrival day, (6) had burns and unclassified injuries, and (7) received neither FAST nor CT.

Eligible patients were divided into (1) those who underwent only FAST or FAST followed by CT as the initial FAST group and (2) those who underwent only CT as the initial CT group.

Variables and outcomes

For this study, we examined the following patient background characteristics: age, sex, Trauma and Injury Severity Score (TRISS),14 type of trauma (blunt or penetrating), cause of trauma (accident, suicide, violence and others) and admission year. TRISS combines age, type of trauma, trauma physiological (revised trauma score) and anatomical (injury severity score) components, and calculated prediction of survival. The primary outcome in this study was in-hospital mortality, and the secondary outcome was length of stay.

Statistical analysis

Continuous variables were presented as medians and IQRs. Categorical variables were presented as numbers and percentages. Baseline characteristics and crude outcomes were compared using the Mann-Whitney U test for skewed distribution of continuous variables and the χ2 test or Fisher’s exact test for categorical variables between the initial FAST and initial CT groups.

We performed multivariable logistic regression analysis for in-hospital mortality and multivariable linear regression for length of stay using the following variables: age, sex, TRISS, type of trauma, cause of trauma and year. We used a generalised estimating equation with an independence correlation structure to account for institution clustering.15

The initial FAST group was further divided into patients with FAST followed by CT and patients with FAST only. Another multivariable linear regression analysis was performed to compare the length of stay between the FAST followed by CT group and the FAST only group, with adjustment for age, sex, TRISS, type of trauma, cause of trauma and year.

TRISS was categorised into quartiles: very low (0–0.92), low (0.93–0.97), high (0.98–0.99) and very high (>0.99).

ORs and associated 95% CIs were calculated for in-hospital mortality and differences, and 95% CIs were calculated for differences in length of stay.

We applied a multiple imputation method to the multivariable analyses because there were missing values for sex, TRISS, cause of trauma, year, in-hospital mortality and length of stay. All of these variables and type of trauma were used to predict imputation. We applied five imputed datasets, using the multivariate imputation for missing data and combined the estimates from the five imputed datasets using Rubin's rules to obtain combined imputation estimates and SEs.16 17

The two-sided significance level for all tests was p<0.05. All analyses were performed using IBM SPSS V.25.

Results

A total of 9942 patients met the inclusion criteria during the study period. Of these, 8558 (86%) were allocated to the initial FAST group, and 1384 (14%) were allocated to the initial CT group (figure 1). In the initial FAST group, 8045 (94.0%) patients underwent FAST followed by CT and 513 (6.0%) underwent FAST only.

Figure 1

Study flow diagram of included patients. FAST, focused assessment with sonography in trauma.

The proportions of the initial CT group were 3.0%, 10.0%, 15.6%, 15.9% and 16.1% in 2004–2006, 2007–2009, 2010–2012, 2013–2015 and 2016, respectively (figure 2).

Figure 2

Proportions of the initial CT group by year ranges.

Baseline characteristics before multiple imputation are shown in table 1.

Table 1

Eligible patients’ baseline characteristics

The median age was lower in the initial FAST group than in the initial CT group. The types of trauma and causes of trauma were significantly different between the groups. There were no significant differences in sex and TRISS between the initial FAST and initial CT groups.

The numbers (percentages) of patients with missing data are also shown in table 1. The percentages of patients with missing data on discharge status (alive or dead) were 5.8% in the initial FAST group and 3.9% in the initial CT group. The percentages of patients with missing data on length of stay were 5.0% in the initial FAST group and 2.8% in the initial CT group.

Crude in-hospital mortality and length of stay are shown in table 2.

Table 2

Crude mortality, hospital stay and time from arrival to CT among patients undergoing initial FAST and initial CT

Crude in-hospital mortality was 5.8% in the initial FAST group and 3.9% in the initial CT group. The median length of stay was 15 days in the initial FAST group and 12 days in the initial CT group. The median time from arrival to CT was 31 min in the initial FAST group and 30 min in the initial CT group.

Multivariable logistic regression showed no significant difference in in-hospital mortality between the initial FAST and initial CT groups (OR 1.37, 95% CI 0.94 to 1.99, p=0.10) (table 3).

Table 3

Multivariable logistic regression for in-hospital mortality fitted with generalised estimating equation

Multivariable linear regression showed a significantly longer length of stay in the initial FAST group than in the initial CT group (difference: 3.5 days; 95% CI 1.0 to 5.9, p=0.006) (table 4).

Table 4

Multivariable linear regression for length of stay (days) fitted with a generalised estimating equation

Multivariable linear regression analysis revealed a significantly longer length of stay in the FAST followed by the CT group than in the FAST only group (difference: 3.1 days; 95% CI 0.5 to 5.7, p=0.02).

Discussion

The present nationwide retrospective cohort study showed no significant differences in in-hospital mortality between the initial FAST and initial CT groups, whereas the initial FAST group showed a significantly longer length of hospital stay than the initial CT group.

Crude in-hospital mortality was higher in the initial FAST group. After adjustment for the years and patient background characteristics, in-hospital mortality showed no significant difference. These results indicate that mortality was affected by the variables included in the multivariable analyses.

Several studies have thrown doubt on the usefulness of FAST.18 19 Previous studies have shown that FAST had a very low sensitivity for detecting blunt intra-abdominal injury.19 20 In 114 haemodynamically stable patients, FAST was positive in 25 patients, with a sensitivity of 22%. In 32 haemodynamically unstable patients, FAST was positive in nine patients, with a sensitivity of 28%.2 Because of the low sensitivity of FAST, it is difficult to detect small intraperitoneal haemorrhage. Furthermore, the main purpose of FAST is to detect free fluid. Because FAST cannot clearly detect organ injury and the origin of haemorrhage, CT is usually performed after FAST. A recent Cochrane review demonstrated that FAST examination did not rule out injuries and must be verified by a reference test such as CT.21 A previous prospective study showed different results for FAST because of rapid access to the operating room. The study included haemodynamically unstable patients and this may have contributed to the results.22 In addition, the previous prospective study did not assess mortality.

The present study also showed that the proportions of patients undergoing initial CT gradually increased between 2004 and 2016. The usefulness of CT in emergency medical settings for trauma has been documented.23 A multicentre observational study showed that CT was associated with a reduction in emergency surgery.23 CT has become more common in trauma care and is now easy to access in most emergency departments.

We excluded children and adult patients who arrived by helicopter or physician-staffed ambulance. The number of children was small, and the clinical characteristics of children tend to differ from those of adults. In addition, the sensitivity of FAST is lower in children than in adults,21 and children may be less likely to undergo CT because of concerns regarding radiation exposure. Regarding adult patients who arrived by helicopter or physician-staffed ambulance, FAST may have been carried out at the scene, and this may have affected physician decisions on the performance of initial CT or FAST in the emergency department.

In the present study, length of stay was longer in the initial FAST group than in the initial CT group. This may be because initial FAST may have led to delayed CT and treatment, resulting in prolonged length of stay. However, there are many possible reasons for the prolonged length of stay that we were unable to measure. The reason for the difference in length of stay remains unclear because the length of stay is affected by many factors.

There are several limitations in this study. First, this was a retrospective study and we evaluated only patients with trauma with proven injuries because of the nature of the data registry system. Second, we excluded patients undergoing neither FAST nor CT because severity in these patients varied. Third, we could not distinguish extended FAST, which incorporates pneumothorax and potential inferior vena cava filling, from standard FAST.24 25 Fourth, the study period was long, and trauma care, ultrasound machine resolution and ultrasound use may have changed during this time. Although we adjusted for the years and several measured confounders in the multivariable regression analyses, unmeasured confounders may have remained. Even though we used a generalised estimating equation model to account for institution clustering, residual confounders in hospital characteristics are possible. For example, FAST has been shown to play an important role in hospitals in which CT examinations or radiologists are not always available.26 However, Japan has the highest number of CT machines per capita, according to the OECD Health Statistics,27 and almost all acute care hospitals in Japan have a CT machine.

In summary, despite a difference in mortality in favour of initial CT, there was no significant difference when confounding variables were taken into account. We found an increased length of stay with initial FAST, the reasons for which remain unclear. Initial FAST may still have a role when access to CT is limited, but CT should be the first-line investigation to detect intra-abdominal injuries in patients with haemodynamically stable torso trauma when available.

Conclusions

We found no differences between the initial FAST and initial CT groups regarding reduced mortality of patients with haemodynamically stable torso trauma. We believe that the usefulness of FAST is becoming unclear in modern clinical settings, and initial CT should be considered in patients with haemodynamically stable torso trauma.

Supplemental material

References

Footnotes

  • Contributors YK conceived this study, analysed the data and drafted the manuscript. HO partially analysed the data and revised the manuscript. HY and HT advised on the design and revised the manuscript. All authors checked and agreed on the final manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The protocol of the present study was approved (number 29-061) by the ethics committee of the Juntendo University Urayasu Hospital.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request.