Analyzing fat embolism syndrome in trauma patients at AIIMS Apex Trauma Center, New Delhi, India.

BACKGROUND: Fat embolism syndrome (FES) is a constellation of symptoms and signs subsequent to orthopedic trauma.
MATERIALS AND METHODS: The clinical profile of FES in the trauma population was studied over 2 years and 8 months.
RESULTS: The incidence of FES among all patients with long bone and pelvic fractures was 0.7% (12). The mean injury severity score was 10.37 (SD 1.69) (range 9-14). The diagnosis of FES was made by clinical and laboratory criteria. Hypoxia was the commonest presentation (92%). The average days of onset of symptoms were 3.5 (SD1.29) days. Management included ventilator support in 75%, average ventilator days being 7.8 (SD 4.08) days. The average ICU stay and hospital stay were 9.1 days and 29.7 days, respectively. A mortality of 8.3% (1) was observed.
CONCLUSION: Fat embolism remains a diagnosis of exclusion and is a clinical dilemma. Clinically apparent FES is unusual and needs high index of suspicion, especially in long bone and pelvic fractures.

INTRODUCTION

Fat embolism is the presence of fat globules in the peripheral circulation and lung parenchyma most often after a long bone fracture or major trauma. The fat embolism syndrome (FES) has more serious manifestation of a multisystem involvement. Clinical onset of symptoms may occur within 12 hrs; but usually patients manifest symptoms 24-72 hrs later. Presentation is variable and no individual symptom is diagnostic of the syndrome. The criterion for diagnosis of FES remains controversial. Some would accept only histological demonstration of fat macroglobules in the organs, while others define the syndrome by clinical abnormalities. Many of the clinical abnormalities are subtle, and must be diligently sought in patients with high suspicion by careful prospective monitoring. Classically patients present with respiratory, neurological and cutaneous manifestations as described by Gurd.[1] Over the years various authors have adapted the Gurd's criteria with slight changes. A retrospective analysis was done to study the incidence and clinical profile of FES in trauma population in our level I trauma center.

MATERIALS AND METHODS

A retrospective case record analysis was done in our level I trauma center, after seeking ethical committee approval medical records of patients from April 2007 to November 2009 in whom FES was diagnosed as per Gurd's criteria were reviewed. To meet the clinical diagnosis of FES, patients needed to demonstrate at least 1 major and 3 minor or 2 major and 2 minor signs. Data collected included demographics, injury severity scores (ISS), clinical presentation, time of onset of symptoms of FES, laboratory investigations, management and outcomes. Comparison between various diagnostic criteria was studied.

RESULTS

There were 1692 patents with long bone and pelvic fracture, admitted during the study period. Fracture femur formed the largest contributor 60.4% (1022), followed by fracture tibia-fibula at 32.6% (552) and pelvic fractures at 7% (118). Twelve cases were identified to have FES as per Gurd's criteria (incidence of 0.7%). Five of 12 patients had multiple bone fracture (42%) and one patient had associated vertebral fracture. The male to female ratio was 3:1. All patients were victims of road traffic accidents with mean ISS of 10.377 (SD ±1.69) with a range of 9-14 [Figure 1]. The average age of patients was 24 years (range 18-28 years). The mean days of onset of symptoms were 3.5 days (SD ±1.29). Four patients presented on the second day (24-48 hrs), three patients each on the fourth (72-96 hrs) and fifth day (96-120 hrs). Unusually, one patient presented postoperatively [Figure 2].

Figure 1

Injury severity score (ISS) distribution for patients with fat embolism syndrome (‘x’ axis: Injury Severity Score, ‘y’ axis: No. of patients)

Figure 2

Onset of symptoms after injury

None of the patients presented with all three major signs of Gurd's criteria. Distribution of major and minor signs as per Gurd's criteria, have been tabulated [Table 1]. Hypocalcemia and hyperbilirubinemia was seen in six patients (50%) and hypoalbuminemia was present in five patients (41.7%). Computerized tomogram (CT) pulmonary angiography was available in six patients with the findings of three (50%) being consistent with the diagnosis of FES. Magnetic resonance imaging (MRI) brain was done in three patients with neurological status alteration and was diagnostic of cerebral fat embolism (CFE) in all three patients. Multiple punctuate, scattered nonconfluent T2 and flair hyperintense lesions were seen in bilateral cerebral hemisphere, basal ganglia, thalamus, pons and cerebellum.

Table 1

Clinical presentation as per Gurd's criteria

The management of these patients was primarily supportive. Steroids were not administered in any of our patients. Early fixation within 12-48 hrs was done in three patients (25%). Nine patients required ventilator support due to respiratory distress, hypoxia and poor Glasgow Coma Score. The average ventilator days were 7.8 (SD 4.08) days and the average stay in the intensive care unit was 9.1 days. The average hospital stay was 29.7 days. One patient died of ARDS and septicemia consequent to fracture femur and subsequent complications.

DISCUSSION

Fat embolises in almost every patient of long bone fracture or major tissue trauma.[2] Gurd's criteria were proposed in 1974 and since then there is no ‘universal criteria’ or ‘gold standard’ for the diagnosis of FES.[3] The incidence of post-traumatic FES has been reported to be as low as 0.2-0.9% in retrospective studies to as high as 35% in prospective studies [Table 2].[123-6] Clinically apparent FES in our study population was 0.7%. Our study population comprised only of young adults as this particular group of patients is more prone for road traffic accidents. There were no pediatric patients diagnosed with FES in our study.

Table 2

Incidence and mortality of fat embolism syndrome in various studies

Major signs

Patients with FES present with a classic triad of respiratory manifestations (95%), cerebral effects (60%) and petechiae (33%).[26] Fifty percent of the patients with FES caused by long bone fractures develop severe hypoxemia and respiratory insufficiency and require mechanical ventilation.[7] Majority of our patients (92%) had respiratory distress and hypoxic presentation. Other differential diagnoses of hypoxia were excluded. Bulger et al noted acute hypoxia in 96% patients of FES.[2]

Neurological signs due to cerebral emboli occur in up to 86% of cases and often occur after the development of respiratory distress.[8] The more common presentation as in our cases (25%) was acute confusional state although focal neurological deficits (hemiplegia, aphasia, apraxia, visual field disturbances, anisocoria) have also been described.[8] None of our patients with neurological deterioration had any residual neurodeficit.

Petechial rash is considered pathognomonic of FES and reportedly present in upto 60% of patients, usually on the conjunctiva, oral mucous membranes and skin folds of the neck and axillae.[9] Despite cutaneous changes being a major criterion of the Gurd's criteria, only one of our patients was reported to have petechial rash with an incidence of 8.3%. This probably could be attributed to the difficulty and delay in identifying rash in the darker skin of the Asian population, its presence for a short duration of time and retrospective nature of the study.

Other than Gurd's there have been attempts at developing other diagnostic criteria, such as those described by Lindeque and Schonfield.[11011] Lindeque based the diagnosis of fat embolism solely on the respiratory status of the patient stating that Gurd's criteria may underdiagnose FES.[10] According to the criteria suggested by Lindeque in our study 11 of 12 patients qualified. Schonfeld quantified the diagnosis of FES [Table 3].[11] As per Schonfeld's criteria 10 of our patients qualify by having a cumulative score of greater than 5 [Table 4]. Schonfeld gave the maximum score to petechiae which was, however, seen in only one of our patients.

Table 3

Lindeque criteria

Table 4

Schonfeld's criteria

Minor signs and laboratory tests

Recovery of fat globules in blood and identification of fat droplets within cells recovered by bronchoalveolar lavage (BAL) has been suggested as a rapid and specific method for establishing the diagnosis of the FES, although of uncertain significance.[1213] Our patients were not tested for fat globules in sputum or BAL. Anemia and thrombocytopenia unexplained by any other cause were observed in nine (75%) of our patients. The cause for hypocalcemia is not well-understood but may result from affinity of plasma free fatty acids (FFA) for calcium or elevated serum lipase.[814] Hypoalbuminemia (present in five patients) has also been suggested due to FFA binding to albumin.[15]

Imaging studies

The chest radiographic appearance of fat embolism is nonspecific.[14] Although chest imaging play miniscule role in diagnosis and management of FES, recognition of the temporal sequence of radiological abnormalities can be a clue in distinguishing pulmonary FES from other pulmonary conditions.[1415] Thoracic CT does not contribute to the regular chest radiographs to aid diagnosis of FES, although it shows earlier, minor infiltrates than ordinary X-ray.[16] However, CT is an insensitive diagnostic modality in FES.[917]

CFE is an uncommon but potentially life-threatening complication of long bone fractures. In our study three patients had neurological deterioration (altered mentation, confusion, disorientation) and all had diagnostic MRI brain findings. Although CT brain findings were not contributory to the diagnosis of CFE, it played a role in excluding other causes of neurological dysfunction. Thus when brain lesions are seen on MRI without head injury the possibility of CFE must be considered. MRI findings in the acute stage of CFE consist of diffuse patchy lesions throughout the brain, most commonly in the white matter and sub cortical grey matter. The lesions were demonstrated by low signal on T2 -weighted images in the cerebellum, cerebrum and brain stem as seen in our patients. T2 -weighted gradient-echo provides useful information on determining the clinical severity of patients with CFE [Figures 3a and 3b].[18] MRI is highly sensitive and specific in detecting encephalic lesions in FES and can detect lesions as small as 2 mm and as early as 4 hrs after trauma.[1920] All three patients with CFE completely recovered without any neurological deficit, over a period of 3 weeks.

Figure 3

Axial T2-weighted images showing hyperintensities in the corpus callosum, bilateral thalami (a) and centrum semiovale (b)

Treatment

Consideration of the differential diagnosis is essential. Other pulmonary pathological conditions were excluded before confirming the diagnosis. Medical care includes adequate hydration, oxygenation, ventilation, stable hemodynamics and blood products as clinically indicated. The use of corticosteroid prophylaxis is controversial, although few studies have shown decrease in the incidence and severity of FES.[81011] No steroids were administered in our patients.

Our centre being a referral center, early fracture fixation was possible in four (33.3%) of our patients. Amongst these, one patient developed post operative FES. This may be attributed to the type of fracture and intraoperative technique of fixation (intramedullary reaming). If patient develops FES following long bone fracture, it is best to delay surgery till patient improves.[21] In the current era of advancement in diagnostic and therapeutic modalities, careful clinical examination teamed with a high degree of suspicion often clinches the diagnosis of FES.

Outcome

Most patients of FES recover completely if adequate supportive treatment and improved nursing is provided. The overall mortality for this condition is 5-15%, with severity of respiratory problems being a close indicator of the risk of death.[22] The mortality in our study population was 8.3%. Bulger et al also reported 7%.[2] The major cause of death in FES patients is the progressive respiratory failure, when the pulmonary lesion evolves to ARDS.[23] In our case the autopsy confirmed ARDS. Mortality rates between 14 and 87% were reported in FES occurred in polytrauma cases.[2425] Early fixation of fractures and modern ICU care has decreased the incidence of FES to be less than 10%.[2627]

CONCLUSIONS

Disparity among the studied diagnostic criteria and the absence of a gold standard for diagnosis of FES makes it an unsolved enigma. Based on our experience from the present study we suggest certain criteria to aid the early diagnosis of this syndrome. 1) High priority indicators are fracture long bone and pelvis, multiple bone fracture, respiratory distress, onset of symptoms in 72 hrs and fever. 2) Low priority indicators: unexplained anemia, thrombocytopenia, tachycardia, hypocalcemia, petechiae, deterioration of neurological status and X-ray chest findings. 3) CFE is self-limiting with no residual neurological deficit. 4) The management primarily remains supportive: modern ICU care, ventilator support and efficient nursing care. Clinical expertise with high degree of suspicion in susceptible trauma population enables diagnosis and appropriate management of this relatively rare but significant syndrome.