Key words trauma - ultrasound - FAST - tertiary trauma survey
Introduction
Patients suffering from severe blunt abdominal trauma are challenging for emergency
room management. These patients need immediate and accurate diagnostic imaging as
well as fast therapeutic action. In accordance with existing trauma guidelines, whole-body
computed tomography (WBCT) represents the gold standard for emergency diagnosis in
the trauma bay [1 ]
[2 ]. WBCT provide a rapid diagnosis tool, which reduces mortality in severely injured
patients [3 ]
[4 ]
[5 ]
[6 ]. Recently, in a large multicenter study, Huber-Wagner et al. demonstrated the potential
benefits of WBCT even in hemodynamically unstable trauma patients with increased survival
due to the speed and high diagnostic quality of modern CT scanners [7 ].
The focused assessment with CT in trauma (FACTT) protocol reduces the number of missed
injuries (MI) in comparison to sonography and conventional radiologic diagnostic imaging
alone [4 ]
[5 ]
[8 ]
[9 ]
[10 ]. However, previous studies reported an incidence of MI between 1.9 – 39 % [11 ] and missed abdominal injuries (MAI) between 0.06 – 15 % [12 ]
[13 ]
[14 ]. Therefore, abdominal follow-up sonography (AFS) during tertiary trauma survey (TTS)
was implemented in international trauma guidelines [15 ]
[16 ]
[17 ]
[18 ]
[19 ]
[20 ]
[21 ]
[22 ]. However, there is only poor evidence for AFS as part of TTS, and therefore the
role of AFS is still unclear. Critics stated that AFS may be too time-consuming and
expensive and its benefit has not yet been sufficiently examined [20 ].
The aim of this study is to analyze trauma patients without abdominal parenchymal
lesions or free intraabdominal fluid in the initial WBCT within emergency room management
in order to assess the impact of AFS on therapy, clinical progress and patient outcome.
Materials and Methods
Study Design
In this retrospective single-center cohort analysis, all patients at an university
hospital and level I trauma center between January 2008 and December 2011 were screened
for trauma room admission, performed WBCT and conducted AFS within 24 hours after
hospital admission as part of the TTS. Patients were checked for inclusion and exclusion
criteria, then clinical and radiographic data were collected and finally descriptive
analysis was performed. The study was approved by the local ethics committee.
Inclusion and exclusion criteria
During the four-year study period, all adult patients who underwent WBCT without evidence
of free intraabdominal fluid or abdominal parenchymal organ lesions, and who also
received an AFS scan within 24 hours after hospital admission during TTS, were included.
The exclusion criteria were pediatric patients (age < 18 years), free intraabdominal
fluid or abdominal parenchymal organ lesions in WBCT ([Fig. 1 ]).
Fig. 1 Presentation of the study design. Abbreviations: WBCT: whole-body computed tomography,
AFS: abdominal follow-up sonography.
Abb. 1 Darstellung des Studiendesigns.
Variables
The patient characteristics included age and gender, length of stay (ICU and hospital),
24-hour mortality, 30-day mortality, and hospital mortality. The incidence of pathological
findings in the AFS as well as secondary bleeding events (e. g. intra-abdominal bleeding
within 24 hours) were recorded.
The prehospital dataset included blood pressure, Glasgow Coma Scale (GCS), cardiopulmonary
resuscitation (CPR), intubation, and the accident mechanism. The emergency room data
record included documentation of injury severity (e. g., abbreviated injury scale
(AIS), injury severity score (ISS), new ISS (NISS)). Moreover, intubation rate, CPR,
and blood transfusion (massive transfusion was defined as more than 9 units of red
blood cells (RBC) were accessed. The ICU dataset included SAPS II, APACHE II and SOFA
for the first 24 hours after admission, length of mechanical ventilation, amount of
administered blood products and the need for organ replacement therapy. Blood chemistry
was registered at the time of initial emergency room management and after ICU admission
(e. g., hemoglobin and lactate level, base excess, pH, partial thromboplastin (PTT)
and prothrombin time (PT) and International Normalized Ratio (INR)). Hemoglobin levels
were also acquired 24 hours after ICU admission in order to detect possible bleeding
situations.
The following were documented as outcome parameters: death within 24 hours after hospital
admission, death during the trauma-associated hospital stay, death within 30 days,
the occurrence of secondary bleeding, the detection of any trauma-associated damage
in the AFS and surgical interventions within 24 hours.
Radiographic data
WBCT is basically defined as a non-enhanced native head CT followed by a contrast-enhanced
thoracic, abdominal and pelvic CT including the whole spine [1 ]. Different multidetector CT scanners and different contrast media injection protocols
are used in the clinical routine. Mostly the traditional segmental approach (the thorax
is scanned in the angiographic phase and the abdomen is scanned in the portal venous
phase) or the continuous angiographic approach are performed [23 ]. A good general overview of modern CT diagnostic methods in major trauma management
is given by Huber-Wagner et al. [24 ].
At our institution two different scanners, scanning protocols and contrast media injection
protocols were used. Between January 2008 and March 2011, all trauma scans were performed
with a 64-row dual-source MDCT scanner (Sensation, Siemens Healthcare, Forchheim,
Germany) and a monophasic scanning protocol. Since April 2011, all patients have been
investigated with a 40-row sliding gantry computed tomography scanner (Sensation open,
Siemens Healthcare, Forchheim, Germany) and a biphasic injection protocol. In most
cases contrast medium (370 mg iodine/mL; Ultravist 370, Bayer Healthcare, Berlin,
Germany) was administered via an 18 G peripheral access or through central venous
access devices. Body weights, renal clearances, thyroid functions and history of possible
contrast media reactions were not known in the emergency setting and had no impact
on the examinations.
Data sources and data management
Clinical data was extracted from the patient data management system (PDMS) ICUData
(IMESO® GmbH, Germany) and subsequently anonymized. Statistical analyses were performed with
IBM® SPSS® Statistics (Version 22.0.0.0, IBM, Armonk, USA).
Statistical analysis
Due to the study design, statistical analysis was performed as a retrospective description
of the study cohort. Clinical data was reviewed on plausibility and descriptive analyses
were performed. Baseline characteristics were expressed as mean ± standard deviation
(SD) in normally distributed data and as median ± interquartile range in not normally
distributed data. For descriptive analysis of absolute and relative frequencies, contingency
tables were used. In the frequency analysis, the 95 % confidence intervals (CI) were
calculated using Monte-Carlo statistics with bootstrap 1000 samples. If an event has
not occurred, CI was assumed for the calculation that the next event would have been
positive (e. g., at 100 negative events (0/100), the estimated upper confidence interval
(EUCI) was calculated as one positive of 101 events (1/101).
Results
During the study period, we identified 1129 emergency room patients who underwent
WBCT examination. 73 pediatric patients, 661 patients with a negative WBCT result
or missing AFS as well as 79 patients with an abdominal injury (n = 79) were excluded
from the following investigation. The study cohort included 316 patients with an inconspicuous
abdominal WBCT scan and an AFS scan performed within 24 hours after trauma.
Patients characteristics
The 316 included patients had a mean age of 43 ± 19 years, and 216 (68.4 %) were male.
An initial GCS below 9 was present in 32 patients (10 %). The mean ISS was 10 ± 8
([Fig. 2 ]). 74 patients (23.4 %) were classified a>s multiple trauma (“polytrauma”, ISS ≥ 16).
The APACHE II, SOFA, and SAPS II were 11 ± 8, 3 ± 3, and 17 ± 11, respectively. The
30-day mortality and overall hospital mortality were 1.8 % and 1.9 %, respectively.
None of the patients died during the first 24 h after hospital admission.
Fig. 2 Bar graph presenting the injury severity score (ISS) of the study population.
Abb. 2 Balkendiagramm zur Veranschaulichung des „injury severity score“ (ISS) der Studienpopulation.
[Table 1 ] shows the demographic characteristics and laboratory and clinical findings reported
for prehospital, emergency room, and ICU. The injury patterns were as follows: 16.1 %
(51) injuries affected the face, 26.4 % (83) the head or neck, 46.7 % (148) the chest,
11 % (35) the lower spine, 48.6 % (154) the extremities or pelvic girdle. External
injuries were present in 46.4 % of cases. Detailed characterization of the subpopulation
of multiple trauma patients featuring an ISS> 15 is presented in [Table 2 ].
Table 1
Study population.
Tab. 1 Studienpopulation.
n (%)
mean
SD
median
Q25
Q75
min
max
gender male
216 (68.4)
age
years
316
43.1
19.01
41.0
27.0
56.0
18.0
95.0
hospital stay
days
316
12.3
12.1
8.3
2.9
17.7
0.35
92.0
ICU stay
days
316
3.9
7.9
1.2
0.8
3.1
0.07
91.9
time from trauma room to US follow-up
hours
316
9.1
10.6
6.7
4.0
10.8
0.4
23.9
pre-hospital
CPR
2 (0.6)
GCS < 8
32 (10.2)
endotracheal intubation
37 (11.7)
trauma room
PTT
s
316
29
6
28
26
31
8
81
INR
316
1.2
0.4
1.1
1.0
1.2
0.9
6.0
BE
316
–2.7
2.9
–2.4
–4.5
–0.3
–12.4
2.6
arterial lactate
316
1.8
1.4
1.5
1.0
2.2
0
9.5
hemoglobin
mg/dl
316
137.9
20.4
141.0
126.0
153.0
68.0
186.0
pH
316
7.341
0.065
7.347
7.313
7.383
7.084
7.476
hemoglobin < 90 mg/dl (n = 309)
16 (5.17)
red blood cell transfusion (n = 313)
0 ml
297 (94.9)
300 ml
1 (0.32)
600 ml
4 (1.28)
900 ml
3 (0.96)
1200 ml
2 (0.64)
1500 ml
2 (0.64)
2400 ml
4 (1.28)
endotracheal intubation
23 (7.3)
CPR
0
time from hospital admission to WBCT
minutes
316
36
22
30
23
44
5
159
ISS
316
10
8
9
3
14
0
43
NISS
316
13
11
11
3
17
0
59
ICU
red blood cell transfusion (in transfused patients)
ml
16
1359
1162
900
600
1500
200
6300
red blood cell transfusion (overall)
ml
316
249
722
0
0
0
0
6300
arterial lactate
316
1.8
1.1
1.5
1.0
2.2
0.5
7.9
hemoglobin
mg/dl
316
126
22
128
112
143
67
174
hemoglobin 24 h
mg/dl
316
118
22
121
100
134
64
172
BE
316
–2.5
2.9
–2.2
–4.0
–0.8
–24.3
6.5
pH
316
7.347
0.06
7.351
7.307
7.382
7.175
7.570
PTT
s
316
31
8
30
27
34
20
92
INR
316
1.2
0.3
1.1
1.0
1.2
0.8
3.3
APACHE II
316
11
8
8
5
13
0
37
SOFA
316
3
3
1
1
3
0
13
SAPS II
316
17
11
15
9
22
0
52
duration of invasive ventilation
days
316
4.7
7.9
1.0
0.4
4.6
0.04
32.7
secondary intraabdominal bleeding
0
need for intraabdominal surgical intervention (24 h)
0
need for intraabdominal surgical intervention (30 d)
0
intraabdominal compartment syndrome
0
positive AFS
3 (0.9)
24-h mortality
0
30-day mortality
4 (1.8)
hospital mortality
6 (1.9)
Abbreviations: ICU: Intensive Care Unit, CPR: Cardiopulmonary Resuscitation; GCS:
Glasgow Coma Scale; PTT: Prothromboplastin Time; INR: Internal Normalized Ratio; BE:
Base Excess; ISS: Injury Severety Score; NISS: New Injury Severity Score; APACHE II:
Acute Physiology And Chronic Health Evaluation; SOFA: Sequential Organ Failure Assessment
score; SAPS II: Simplified Acute Physiology Score; AFS: Abdominal Follow-Up Sonography
Table 2
Study characteristics of patients ISS > 15.
Tab. 2 Studiencharakteristiken der Patienten mit einem ISS ≥ 15.
n (%)
mean
SD
median
Q25
Q75
min
max
gender male
57 (77)
age
years
74
46.1
20.65
45
24
62
18
83
ISS
74
23
6
22
17
27
16
43
NISS
74
27
9
24
22
29
16
59
APACHE II
74
16
9
14
8
24
2
37
SOFA
74
5
4
4
1
8
0
13
SAPS II
74
23
10
22
17
28
7
52
secondary intraabdominal bleeding
0
need for intraabdominal surgical intervention (24 h)
0
need for intraabdominal surgical intervention (30 d)
0
intraabdominal compartment syndrome
0
positive AFS
2 (2.7)
24-h mortality
0
30-day mortality
2 (3.3)
hospital mortality
4 (5.4)
Abbreviations: ISS: Injury Severity Score; NISS: New Injury Severity Score; APACHE
II: Acute Physiology And Chronic Health Evaluation; SOFA: Sequential Organ Failure
Assessment score; SAPS II: Simplified Acute Physiology Score; AFS: Abdominal Follow-Up
Sonography
Follow-up abdominal sonography
All 316 included patients with an inconspicuous WBCT scan (absence of free intraabdominal
fluid or abdominal parenchymal organ lesions) did not suffer from secondary intraabdominal
bleeding or intraabdominal compartment syndrome or need intra-abdominal surgical or
interventional procedures during the study period. A positive AFS scan was reported
in 3 patients (0.9 %). Patient 1 (age 29 years, ISS 29, NISS 29) presented a limited
amount of free intraabdominal fluid in the Morrison pouch, ventral to the liver, in
the left lower abdomen around the sigmoid colon; patient 2 (age 36 years, ISS 14,
NISS 22) showed a small anechoic signal inferior to the liver reconcilable with free
intraabdominal fluid; and patient 3 (age 18 years, ISS 24, NISS 29) showed limited
amounts of free intraabdominal fluid at different locations (Douglas cavity, beside
the spleen and the liver and paracolic). All three patients had stable hemodynamic
and respiratory status during the study period and none of the conspicuous AFS scans
led to further treatment or surgical or interventional procedures. [Table 3 ] describes the examination results of patients with a positive AFS scan and patients
who died during the hospital stay. None (0.0 %, EUCI < 1.3 %) of these patients died
during the first 24 h after hospital admission.
Table 3
Cases of death.
Tab. 3 Todesfälle.
age
ISS
NISS
day of death after admission
cause of death
55
9
18
8
hypoxic brain damage after aspiration and CPR
80
18
27
12
pneumogenic sepsis with MOF after stabilized subdural hematoma
77
22
27
4
subarachnoid hemorrhage with infaust prognosis
83
9
27
5
cardiac and pulmonary failure after massive cerebral trauma
83
34
34
91
pneumogenic sepsis with acute renal failure after pulmonary contusion and critical
illness myopathy and neuropathy
74
17
27
31
sepsis with MOF and cholecystitis after subdural hematoma
Abbreviations: CPR: Cardiopulmonary Resuscitation; MOF: Multiple Organ Failure.
Hospital mortality
Six (1.9 %, CI 0.6 – 3.5 %) patients died during their hospital stay, but none of
these patients suffered from intraabdominal bleeding or organ lesion within the 24
hours after trauma ([Table 3 ]). Only one (0.3 %, CI 0.0 – 1.0 %) patient (age 62 years, ISS 22, NISS 22) suffered
from secondary bleeding, which was localized in the extraperitoneal fat tissue and
caused by minor bleeding of the pelvic vasculature, within the first 24 hours after
hospital admission. In this patient AFS did not show free intraabdominal fluid or
parenchymal organ lesion. Subsequently, this patient needed surgical treatment of
his subcutaneous hematoma on day 16 after trauma.
Discussion
AFS during tertiary trauma survey (TTS) was implemented in international trauma guidelines
to reduce the rate of missed abdominal injuries [15 ]
[16 ]
[17 ]
[18 ]
[19 ]
[20 ]
[21 ]
[22 ]. Nonetheless, there is only poor evidence for the need for AFS as part of TTS, and
the role of AFS is still unclear. Therefore, the aim of our study was to assess the impact
of AFS within TTS in patients without parenchymal organ lesion or free intraabdominal
fluid in the initial WBCT examination.
In this retrospective study with 316 patients, AFS, which was performed within 24
hours after WBCT, did not provide any additional information and had no impact on
further treatment decisions. Summarized it can be noted:
First, we investigated a well-defined study population. Compared to the results of
the trauma register of the German society of trauma surgery [21 ], our study cohort had equal patient characteristics concerning gender and age (male:
70 % vs. 68 %, age: 47 years vs. 43 years) but showed different results for in-hospital
mortality (10 % vs. 2 %), mean ISS (17 vs. 10), and ISS ≥ 16 (48 % vs. 23 %). In line
with other parameters (e. g., length of stay in the hospital: 16 d vs. 13 d, ICU length
of stay: 7 d vs. 4 d), these findings indicate a more severely injured patient collective
described in the trauma register. Nevertheless, these results should be interpreted
with caution, because the low mean ISS in our study is certainly caused by the study
design with the exclusion of patients with parenchymal organ lesions or free intraabdominal
fluid on the one hand, and on the other hand by the generous indication for trauma
team activation defined by the German guideline on the treatment of severely injured
patients ([Table 4 ]) [1 ]. Moreover, this might also explain the varying percentages of severe head (11.4 %),
chest (27.8 %) and extremity (16.1 %) injuries with an AIS ≥ 3 compared to the German
Society of Trauma Surgery registry. It is also well known that the sensitivity of
injury severity prediction by a prehospital EMS provider is low [22 ]
[25 ]. Additionally, it should be kept in mind that trauma scores (e. g., AIS, ISS, NISS)
could only be estimated after completion of sufficient diagnostics.
Table 4
Criteria for trauma team activation based on the recommendations of the German Society
of Trauma Surgery [1 ].
Tab. 4 Kriterien der Trauma-Team Aktivierung gemäß der Empfehlungen der Deutschen Gesellschaft
für Unfallchirurgie.
mechanism of trauma
high speed trauma (> 80 km/h)
fall from a height greater than 3 meters
passenger ejection out of a vehicle
death of another passenger
impaction of passenger
burial
pattern of injuries
multiple trauma (ISS> 15)
trauma to the head with disturbance of consciousness
trauma to the thorax with signs of instability
trauma with perforation of the thorax and abdomen
blunt abdominal trauma
trauma to the pelvis with signs of instability
fracture of more than one proximal bone
amputation of a proximal limb
vital signs
GCS < 14
systolic blood pressure < 80 mmHg after trauma
breathing frequency < 10/min or > 29/min after trauma
SpO2 < 90 % after trauma
admission from another treatment facility
prior trauma room treatment in another hospital
ICU admission < 96 hours after trauma in another hospital
incomplete diagnostics in another hospital after trauma
Abbreviations: ISS: Injury Severity Score; GCS: Glasgow Coma Scale; SpO2: mixed oxygen
saturation; ICU: Intensive Care Unit
Second, WBCT is associated with a high sensitivity and specificity for the detection
of intra-abdominal lesions after blunt abdominal trauma and is able to shorten the
time between patient arrival and diagnosis [26 ]
[27 ]. Livingston et al. [28 ] demonstrated a negative predictive value of 99 % for WBCT and concluded that there
are no advantages for prolonged surveillance or hospital admission after completion
of a WBCT examination in patients with blunt abdominal trauma. Moreover, Huber-Wagner
et al. [7 ] demonstrated that WBCT is a relevant and safe diagnostic tool even in hemodynamically
unstable patients associated with increased survival even in severely injured patients.
However, Neal et al. [29 ] controversially found a 70 % higher mortality risk in patients who received an abdominal
CT instead of instant laparotomy during initial emergency room management (n = 3.218).
According to the results of Huber-Wagner et al. [4 ]
[6 ] and the recommendations of the German Society of Trauma Surgery, WBCT is performed
in all patients at risk for abdominal injury at our institution.
Third, focused assessment with sonography for trauma (FAST) is confirmed to be a safe,
fast and highly specific diagnostic tool for abdominal screening in blunt trauma patients.
Therefore, FAST is well implemented in current trauma guidelines. However, FAST only
offers a low sensitivity regarding the detection of organ lesions and is also known
as being highly examiner-dependent [30 ]. Moreover, FAST is limited by a sensitivity of 41 % in hemodynamically stable patients
(positive and negative predictive value 94 % and 95 %) and should be reserved for
hemodynamically unstable patients [31 ]. Compared to FAST, AFS is more time-consuming and needs a greater level of experience.
In line with our investigation, Mauer et al. [20 ] stated that AFS is a time- and personnel-consuming diagnostic process which offers
less additional information, often without any therapeutic consequences, but contributed
a total cost of 29 € per patient. Therefore, its usefulness after WBCT has to be strongly
questioned [31 ]
[32 ]. Considering our results and in line with Geyer et al. [32 ], AFS within the TTS after inconspicuous WBCT examination and after exclusion of
extra-abdominal bleeding sources should only be performed if clinical or laboratory
findings in the observation period are indicative for patient deterioration.
Limitations
Due to the fact that the evidence of this algorithm is only based on retrospective
studies, its safety and efficiency have to be investigated in prospective studies.
In addition to its retrospective design, our study has some other limitations: The
results of our retrospective analysis should be interpreted with caution because our
study population represents only a subgroup of patients suffering from trauma. Therefore,
the results and the derived recommendations cannot automatically be transferred to
other patient collectives (e. g., patients with more severe trauma or parenchymal
organ lesions). Also a higher number of cases and an analysis of other patient groups
and subgroups will be necessary to answer the question regarding the need for AFS
after trauma. Moreover, the results may be biased by the fact that only 30 % of all
patients with a negative WBCT result regarding parenchymal organ lesion or free intra-abdominal
fluid could be included in the study because an AFS scan was missing in 661 patients.
It is to be expected that AFS frequently was not performed in hemodynamically stable
patients with minor injuries without abdominal or pelvic pain or any other symptoms.
One well-known general limitation is that AFS is highly dependent on operator skills,
variability and experience, which certainly might have an impact on further diagnostics
and therapy. Moreover the clinical status of intubated, demented or mentally impaired
patients is difficult to evaluate and therefore clinical worsening might be overlooked
by pure analysis of laboratory parameters. For this reason, the indication for AFS
should be made more generously. Last of all, different CT scanners, scanning protocols
and contrast media injection protocols were used in our study population and might
therefore influence the diagnostic accuracy of WBCT. However, the multiple studies
in the field of trauma scanning revealed high diagnostic accuracy independent of the
CT scanner and contrast media injection protocol being used.
Conclusion
In conclusion, in this retrospective study, AFS as part of the TTS did not show additional
benefits and had no impact on further treatment in patients without abdominal parenchymal
organ lesions or free intraabdominal fluid in the initial WBCT examination. We conclude
that AFS is not routinely required but should be performed if clinical or laboratory
parameters require fast and noninvasive re-evaluation of the trauma patient.