Sternal fractures in a level III trauma intensive care unit

1 University of Sao Paulo (USP), Faculty of Medicine, Department of Surgery Division of General Surgery and Trauma, Sao Paulo, SP, Brazil. 2 University of Sao Paulo (USP), Faculty of Medicine, Department of Anaesthesiology, Sao Paulo, SP, Brazil. Pereira Sternal fractures in a level III trauma intensive care unit. 2 Rev Col Bras Cir 46(1):e2059 This division becomes relevant because fractures in certain areas of the sternum are more related to specific associated lesions4. Remarkably, manubrium fractures are associated with thoracic spine, scapular, and TBI lesions, while Part III fractures are more related to lumbar spine fractures. Most sternum fractures are stable and without obvious misalignments, and have conservative management10. Surgical options may be considered in patients with unstable sternal fractures, unsatisfactory pain control despite optimized clinical treatment, or difficulty in ventilation weaning attributed to fracture. A strategy of late surgical approach may be useful in cases of patients with chronic pain, no fracture consolidation, or aesthetic reasons11. However, in selected groups of patients, such as those with pericardial effusion and possible vascular lesions due to sternum trauma, emergency surgical stabilization is indicated.


R
ibcage lesions are a common type of injury in trauma patients 1 and their presence is related to morbimortality increase, with rates reaching 60% in this population 2 .Motor vehicle accident is the most common trauma mechanism and closed chest trauma is the predominant type of injury 1 .Within this subgroup of patients, sternum fractures were considered uncommon, with incidence ranging from 0.3% to 3.7% 3 .However, this incidence has been increasing in recent decades, fact which may be related to the increased use of seat belts 4 .
In this context, computed tomography (CT) has increased diagnostic accuracy, since up to 94% of all sternum fractures are not diagnosed in initial chest X-ray examination 5 .The use of wholebody CT protocols allows the accurate diagnosis of associated lesions and other complications 6 .
The most common associated lesions in these patients are costal arch fractures, traumatic brain injury (TBI), spinal injuries, and pulmonary contusions 4 .Data from the United States show that mortality rate is low (2.4%) in patients with isolated sternal fracture.However, this rate can increase up to 8.8% in the presence of other associated lesions 7 .In this population, the independent risk factors associated with mortality are thoracic lesion, pulmonary contusions, acute respiratory distress syndrome (ARDS), age, and days of mechanical ventilation.
Literature describes a topographic division of the sternum in four distinct anatomical regions, a specific subdivision for manubrium fractures 4,8,9 (types A, B, and C fractures, depending on the generated instability -sagittal, rotational, or combined, respectively), and the sternal body, divided into these portions: upper (Part I), mid (Part II), and distal, including xiphoid process (Part III).

Original Article
Sternal fractures in a level III trauma intensive care unit.

Pereira
Sternal fractures in a level III trauma intensive care unit.Most sternum fractures are stable and without obvious misalignments, and have conservative management 10 .Surgical options may be considered in patients with unstable sternal fractures, unsatisfactory pain control despite optimized clinical treatment, or difficulty in ventilation weaning attributed to fracture.A strategy of late surgical approach may be useful in cases of patients with chronic pain, no fracture consolidation, or aesthetic reasons 11 .However, in selected groups of patients, such as those with pericardial effusion and possible vascular lesions due to sternum trauma, emergency surgical stabilization is indicated.

This work was approved by the Institutional
Ethics Committee with the following reference number 63077116.6.0000.0068.Most fractures affected the sternal manubrium (62%), followed by the body (30%) and sternoclavicular junction (8%).Only three patients (23%) had unstable chest attributed to sternal fracture, and two underwent surgical fixation of the fracture.Both patients had satisfactory outcomes at hospital discharge (Table 2).3).

DISCUSSION
There is a shortage of national studies on the epidemiology of sternal fractures in critically ill trauma victims.In our series, the prevalence of sternum fractures was 0.9% of total trauma, in agreement with international literature in which prevalence ranges from 0.3% to 3.7% 3 .Our data did not allow us to evaluate possible historical increase trends in the incidence of these fractures.This increased incidence in the last decades described in foreign literature is due to both a greater diagnostic accuracy and to the mandatory use of three-point seat belt for drivers and front passengers, which causes a point of greater pressure in the anterior chest region during deceleration.
The epidemiological characteristics of our series were also in line with international literature, in which most lesions are of sternal body, with motor vehicle accidents as the most common trauma mechanism.However, a considerable portion of our patients had other high-energy trauma mechanisms as causal agent, notably fall from a height and trampling.This reinforces that such mechanisms should not be underestimated as risk factors for traumatic lesion of the sternum, especially given the high prevalence of this trauma mechanism, both in developed and developing countries 12,13 .
Considering that our study had as inclusion criteria only patients admitted to ICU, severity indexes and prognostic scores such as SAPS III, ISS, and NISS were higher than those found in literature.
For the same reason, mortality in our population was higher than in other series.However, most of the deaths occurred in patients with other associated lesions, which had a direct contribution to the cause of death.In fact, only in one patient, the cause of death was directly attributed to sternal fracture.
Therefore, we consider that sternum fractures are a severity marker in patients admitted to ICU.By highenergy trauma mechanism, these fractures are most of the time associated with other severe lesions, such as TBI and pelvic and abdominal traumas.For a very selected population of patients with sternal trauma, especially those with clinical instability of the fracture and unstable chest, surgical fixation, although little studied in clinical trials, seems to bring benefits to these patients, such as facilitating the withdrawal of invasive mechanical ventilation, the main benefit.
Our case series presents some limitations: it is a retrospective study, without an individual analysis of the significance of whole-body tomography in our population, which may increase the prevalence of lesions.
The low number of events does not allow any analysis of the association of risk factors or prognostic predictors in our population.As far as we know, this is the first national historical series which evaluates the prevalence of sternal fractures in a population of trauma victims.The presence of sternal fracture and associated lesions may be related to the outcome of these patients.For a very specific group, surgical fixation may be indicated and beneficial.However, further studies on this topic are needed.

2
Rev Col Bras Cir 46(1):e2059This division becomes relevant because fractures in certain areas of the sternum are more related to specific associated lesions4 .Remarkably, manubrium fractures are associated with thoracic spine, scapular, and TBI lesions, while Part III fractures are more related to lumbar spine fractures.
We conducted a retrospective analysis of adult patients admitted to Intensive Care Unit (ICU) of a Level III trauma center in Sao Paulo, Brazil, from January 2012 to April 2016.This observational study had the objective of describing epidemiology, anatomical characteristics, and outcomes of patients with sternum fracture in this population.Continuous variables were described as median and interquartile range (IQR) or mean and standard deviation, as appropriate.Categorical data were expressed as percentages and absolute values.Due to the descriptive epidemiological nature of this study, no other statistical analysis was performed.
Three patients with sternal fracture died during observation period, two due to nonrespiratory complications.One of these patients had a diagnosis of brain death on the 12th day of ICU admission due to TBI.One patient died nine days after admission due to anoxic-ischemic encephalopathy after cardiorespiratory arrest at the scene.The third patient presented retrosternal hematoma caused by sternal fracture and died on the first day of hospitalization due to shock and respiratory failure (Table

Table 3 .
Patients with fatal lesions.