- Email: [email protected]
Fat Embolism Syndrome and Pulmonary Microvascular Cytology
Fat Embolism Syndrome and Pulmonary Microvascular Cytology· Xavier CasteUa M.D., F.C.C.P.; jordi \blles M.D.; Maria A. Cabezuelo M.D.; Rafael Fernandez M.D.; and Antoni Artigas M.D. Pulmonary microvascular cytology consists of analysis of capillary blood sampled while a Swan-Gaoz catheter is in the wedge position. This technique has proved to be useful in the diagnosis of Iymphangitic spread of carcinoma in the lungs and there are case reports of their use in amniotic fluid embolism. Its usefulness in diagnosing fat embolism syndrome has been shown only rarely. We report a new case in which pulmonary microvascular cytologic study allowed a definite diagnosis of fat embolism syndrome. We suggest obtaining routinely samples of capillary blood when a pulmonary catheter is in place and fat embolism is suspected on a clinical basis. (Cheat 1992; 101:1716-11)
=
=
ALT alanine aminotransferase; APTT activated partial thromboplastin time; AST aspartate aminotransferase
=
P
ulmonary dysfunction following long bone or pelvic fracture is often attributed to fat embolism syndrome. Diagnosis of this syndrome is based on the clinical triad of hypoxemia with new lung infiltrates, U neurologic dysfunction: and petechiae with alterations of the coagulation system' appearing 72 h after bone injury. Other diagnostic possibilities include measurement of serum lipase activity,' fat globules in urine, and other speculative laboratory tests. e.7 Recently Chastre et ale have suggested that bronchoalveolar lavage (BAL) could be an accurate diagnostic procedure in this syndrome. Many authors have shown that a Swan-Ganz catheter may be a diagnostic aid in patients with suspected fat embolism.·· I • A rise in pulmonary artery pressure, pulmonary vascular resistance, and wedge pressure as well as a decrease in arterial oxygen tension could suggest fat embolism in an adequate clinical context. Masson and Ruggieri" introduced the concept of pulmonary microvascular cytology for analyzing samples of blood obtained while the Swan-Ganz catheter is wedged. They reported six patients with suspected fat embolism in whom this technique detected fat globules in capillary blood samples." Since the publication of this important article, we are unaware of other reports in which pulmonary microvascular cytologic study has been useful in diagnosing fat embolism. We would like to add a new case diagnosed using this simple and specific technique. CASE REPORT
A 23-year-old woman was admitted to the hospital after severe trauma from a car accident. At the time of admission she showed normal vital signs except a heart rate of 140 beats/min (bpm); neurologic examination demonstrated a Glasgow coma score of 15. Trauma to both legs was evident, with fractures in right femur, right tibia, and left patella. A diagnostic peritoneal lavage was positive and she was taken to the operating room where two hepatic *From the Intensive Care Medicine (Drs. Castella, Valles, Fernandez, and Artigas) and Pathology (Dr. Cabezuelo) Services, Hospital de Sabadell, Sabadell, Spain.
1710
lacerations were sutured. The patella fracture was repaired and the fractures of the right leg were fixed. On the second hospital day, the patient showed tachypnea and she became cyanotic. Her temperature rose to 3lfC with progressive restlessness. Arterial blood gas values while breathing oxygen at Flo. of0.5 were as follows: pH, 7.45; PaO., 50 mm Hg; and PaCO., 41 mm Hg. A chest roentgenogram revealed diffuse alveolar infiltrates in both lungs. An electrocardiogram showed sinus tachycardia at 120 bpm. The patient was transferred to the intensive care unit (ICU) after tracheal intubation. On ICU admission, her blood pressure was 95150 mm Hg, pulse was 120 bpm, and temperature was 38°C. Physical examination revealed cyanosis, conjunctival petechiae, and diffuse inspiratory rales. A new chest roentgenogram showed diffuse alveolar in61trates with a preference for lower lung fields. Arterial blood gas values were as follows: PaO., 93 mm Hg; PaCO., 35 mm Hg while breathing 100 percent oxygen. Laboratory tests revealed the foUowing: white blood cell count, 11,7001cu mm; hematocrit, 32 percent; hemoglobin, 11 gldl; platelet count, 77,OOOIcu mm; prothrombin time, 65 percent; activated partial thromboplastin time (APnj 40 s (control, 36 s); fibrinogen, 685 mgldl; and 6brin(ogen) degradation products above 4,000 nglml; total billirubin was 3.4 mgldl with a direct component of 2.3 mgldl; creatine phosphokinase, 1,064 UIL with a MB fraction of27 UIL; serum alanine aminotransferase (ALl') was 75 UIL and serum aspartate transaminase (Am was 55 UIL. A pulmonary artery catheter was inserted, revealing a cardiac output of 6.3 Umin, pulmonary arterial pressure of 41/22 (31) mm Hg, and a pulmonary capillary wedge pressure of 10 mm Hg. Pulmonary and wedge position were con6rmed using published criteria consisting in analysis ofhemodynamic pressures, waveforms, and blood gas values. With the diagnosis ofadult respiratory distress syndrome (ARDS) probably secondary to fat embolism, we took nonheparinized samples of microvascular blood while the pulmonary artery catheter was in the wedge position. We rejected the first 10 ml of blood assuming that this amount represents "dead space" coming from terminal pulmonary arteries. These samples were immediately transferred to the pathology laboratory where they were centrifuged to separate the cellular and noncellular components of the 8uid. After this we 6xed the cells using 96 percent alcohol or letting them dry at room temperature. Preparations were stained using the Papanicolaou method in the first case or May-Grunwald-Giemsa and Sudan black method otherwise. Microscopic analysis of these samples with the May-GrunwaldGiemsa stain showed large numbers of globules (presumably fat) in a background of hemolyzed blood (Fig 1 and 2). These globules showed a positive pigmentation with the lipid-specific stain Sudan black supporting the initial assumption offat content in the globules (Fig 3). The patient was treated using mechanical ventilation with .:~
~
.'
,"(.l....~,
,. .' .'\41~~ .. .
..
.. '.
~.......
,,"
'\', .....
.•.." :.:
..
'
~ ~:.
""'. -.; . ""-:-".~' .~. > ......".' <.. ~.'~ ...... -.;. ~·r~"' . '·'\":' ·~A.·.~.· _., -,~. . ' ..... '¥....~.:-.\-~,. ~ -r~, ~ .,.'If.1II
. \... ~ .., ._ ..• ~ .._ - "V"; """., , ~ .... ~. • . _ .. ' ~iI' ...trIId...., .•.. ~.. . , ' • ., *':
.:c
.. I.
~""": '_.'.'-,.
~.........
'~".
...::
..
'or
~'. "
....
..
,".
·
~
...
!). III• • #",_,.:..-
.: • • • . . .
.....
•
III
•
.. ........ ...-
•..'
.."
. ,....
,....
_
'~~., •
.....
.'
4
,,"
"
WO'
.
~".
.
.•. ...-
"-..
:'.,~
' •• -
..
~~, .•'"
'i:I... ,.,. ,..
· .' l.·~·. :', .....~ t .... t;~ .. I ' • .•• ......... • .-:J ~: ~''P'T
• •",:,
""~' _ ..-' "'\'."". '.
. ~ ._ "'~~' ,., ~,.' ~.~ ..;.. ~~"'7 .,'d '(" .':'
,.. '''-..\ . ' - "'".' "'.".. ;" ....• '~." •. ,"t'.:.. ;' ~'~"\." •
,
.t
~1"""
,0
~
-,
•
• •11
•
\.~ •
•
'~
.•. .~ -~
. .t
...' ~ '* ;;~;1. .,~ t .....'-* ~., ., .' ~~z·", ""'S ~" . .f ......... .,..~....,. .... ~ ., :.'-"'~ T ..... " . ". J-.' ) " . 't . 1~:t'" • '
'1" _
....
.. ~
'..... . .~
•. ~
'.'
'
#f'
•
t
.:...... OIl.
.(
,.
;...
....
'.
:-,:-:-", ..... •
FIGURE 1. Pulmonary capillary sample from the patient showing globules (presumably fat) from the blood clot (May-GrunwaldGiemsa stain, original magnification X 40). Fat EmboIlsm Syndrome (castella st eI)
Swan-Ganz catheter. Despite this suggestive accuracy, we are unaware of other reports using pulmonary microvascular cytology in this syndrome. Our case corroborates the usefulness of a technique that we think is as specific as lung biopsy. From our experience, we would suggest that treatment of patients with suspected fat embolism should routinely include pulmonary artery catheterization and capillary blood sampling. REFERENCES
FIGURE 2. Closer view of Figure 2 focused on the globules (MayGrunwald-Giemsa stain, magnification x 200). positive end-expiratory pressure (PEEP) and methylprednisolone was added during 24 h. The patient improved and she could be weaned from mechanical ventilation on the sixth day. She was discharged from the ICU on the tenth day. DISCUSSION
Pulmonary microvascular cytology is a new technique that samples capillary lung blood obtained from a wedged SwanGanz catheter." Its usefulness has been shown in diagnosing a few cases of amniotic fluid embolism.'1-l3 Despite some discussion," Masson et al"," proposed capillary blood sampling as a useful diagnostic procedure in lymphangitic carcinomatosis of the lung and established it showed malignant cells in seven of eight samples from patients with Iymphangitic spread of carcinoma in the lungs. '6 Fat embolism is a protean syndrome elusive to diagnosis. 17 Classically it is diagnosed on a clinical basis alone. ,. Schonfeld et al'· have proposed an objective scoring system, but it cannot be used in patients with severe cerebral, thoracic, or abdominal injuries. Detection of fat globules in urine and blood measurement of serum lipase activity have been suggested to be useful in some patients, but these tests have never been evaluated critically. Recently, Chastre et al" have shown that BAL in patients with suspected fat embolism syndrome could be an accurate diagnostic procedure. In six of seven patients with suspected fat embolism, Masson and Ruggieri" could find fat globules in blood obtained from a wedged
FIGURE 3. The globules of Figures 1 and 2 showed a positive pigmentation with the lipid-Specific stain Sudan black supporting the initial assumption of fat content (Sudan black stain, original magnification x 1(0).
1 Evarts C. The fat embolism syndrome: a review. Surg Clin North Am 1970; 50:493-507 2 Dines D, Linscheid R, Didier E. Fat embolism syndrome. Mayo Clin Proc 1972; 47:237-40 3 Jacobson D, Terrence C, Reinmuth 0. The neurologic manifestations offat embolism. Neurology 1986; 36:847-51 4 Keith R, Mahoney L, Garvey M. Disseminated intravascular coagulation: an important feature of the fat embolism syndrome. Can Med Assoc J 1971; 105:74-6 5 Armstrong H, Kuenzig M, Peltier L. Lung lipase levels in normal rats with experimentally produced fat embolism. Proc Soc Exp BioI Med 1967; 124:959-61 6 Renne J, Wuthier R, House E, Cancro J, Hoaglund F. Fat macroglobulimia caused by fractures or total hip replacement. J Bone Joint Surg 1978; 6OA:613-17 7 Lepisto P, Alho A. Diagnostic features offat embolism syndrome. Acta Chir Scand 1975; 141:245-50 8 Chastre J, Fagon J, Soler P, Fichelle A, Dombret M, Huten D, et aI. Bronchoalveolar lavage for rapid diagnosis of the fat embolism syndrome in trauma patients. Ann Intern Med 1990; 113:583-88 9 Dorr L, Merkel C, Mellman M, Klein I. Fat emboli in bilateral total knee arthroplasty: predictive factors for neurologic manifestations. Clin Orthop 1989; 248:112-18 10 Byrick R, Kay J, Mullen J. Capnography is not as sensitive as pulmonary artery pressure monitoring in detecting marrow microembolism. Anesth Analg 1989; 68:94-100 11 Masson R, Ruggieri J. Pulmonary microvascular cytology: a new diagnostic application of the pulmonary artery catheter. Chest 1985; 88:908-14 12 Agia G, Wyatt L, Re E. Pulmonary microvascular cytology: a new diagnostic application of the pulmonary artery catheter. Chest 1986; 90:627-28 13 Vanmaele L, Noppen M, Vincken \v, De Catte L, Huyghens L. Transient left heart failure in amniotic Ouid embolism. Intensive Care Med 1990; 16:269-71 14 Spapen H, Umbrain V. Huyghens L. Case 30-1987: cytology of pulmonary blood samples. N Engl J Med 1988; 318:324 15 Masson R, Lukl P, Krilorian J, Evans G. Case 30-1987: cytology of pulmonary blood samples. N Engl J Med 1988; 318:324-25 16 Masson R, Krikorian J, Lukl P, Evans G, McGrath J. Pulmonary microvascular cytology in the diagnosis of Iymphangitic carcinomatosis. N Engl J Med 1989; 321:71-6 17 Fabian T, Hoots A, Stanford D, Patterson C, Mangiante E. Fat embolism syndrome: prospective evaluation in 92 fracture patients. Crit Care Med 1990; 18:42-6 18 Peltier L. Fat embolism: a perspective. Clin Orthop 1988; 232:263-70 19 Schonfeld S, Ploysonsang Y, DiLisio R, Crissman J, Miller E, Hammerschmidt D, et aI. Fat embolism prophylaxis with corticosteroids: a prospective study in high risk patients. Ann Intern Med 1983; 99:438-43 CHEST I 101 I 6 I JUNE. 1992
1711
=
=
ALT alanine aminotransferase; APTT activated partial thromboplastin time; AST aspartate aminotransferase
=
P
ulmonary dysfunction following long bone or pelvic fracture is often attributed to fat embolism syndrome. Diagnosis of this syndrome is based on the clinical triad of hypoxemia with new lung infiltrates, U neurologic dysfunction: and petechiae with alterations of the coagulation system' appearing 72 h after bone injury. Other diagnostic possibilities include measurement of serum lipase activity,' fat globules in urine, and other speculative laboratory tests. e.7 Recently Chastre et ale have suggested that bronchoalveolar lavage (BAL) could be an accurate diagnostic procedure in this syndrome. Many authors have shown that a Swan-Ganz catheter may be a diagnostic aid in patients with suspected fat embolism.·· I • A rise in pulmonary artery pressure, pulmonary vascular resistance, and wedge pressure as well as a decrease in arterial oxygen tension could suggest fat embolism in an adequate clinical context. Masson and Ruggieri" introduced the concept of pulmonary microvascular cytology for analyzing samples of blood obtained while the Swan-Ganz catheter is wedged. They reported six patients with suspected fat embolism in whom this technique detected fat globules in capillary blood samples." Since the publication of this important article, we are unaware of other reports in which pulmonary microvascular cytologic study has been useful in diagnosing fat embolism. We would like to add a new case diagnosed using this simple and specific technique. CASE REPORT
A 23-year-old woman was admitted to the hospital after severe trauma from a car accident. At the time of admission she showed normal vital signs except a heart rate of 140 beats/min (bpm); neurologic examination demonstrated a Glasgow coma score of 15. Trauma to both legs was evident, with fractures in right femur, right tibia, and left patella. A diagnostic peritoneal lavage was positive and she was taken to the operating room where two hepatic *From the Intensive Care Medicine (Drs. Castella, Valles, Fernandez, and Artigas) and Pathology (Dr. Cabezuelo) Services, Hospital de Sabadell, Sabadell, Spain.
1710
lacerations were sutured. The patella fracture was repaired and the fractures of the right leg were fixed. On the second hospital day, the patient showed tachypnea and she became cyanotic. Her temperature rose to 3lfC with progressive restlessness. Arterial blood gas values while breathing oxygen at Flo. of0.5 were as follows: pH, 7.45; PaO., 50 mm Hg; and PaCO., 41 mm Hg. A chest roentgenogram revealed diffuse alveolar infiltrates in both lungs. An electrocardiogram showed sinus tachycardia at 120 bpm. The patient was transferred to the intensive care unit (ICU) after tracheal intubation. On ICU admission, her blood pressure was 95150 mm Hg, pulse was 120 bpm, and temperature was 38°C. Physical examination revealed cyanosis, conjunctival petechiae, and diffuse inspiratory rales. A new chest roentgenogram showed diffuse alveolar in61trates with a preference for lower lung fields. Arterial blood gas values were as follows: PaO., 93 mm Hg; PaCO., 35 mm Hg while breathing 100 percent oxygen. Laboratory tests revealed the foUowing: white blood cell count, 11,7001cu mm; hematocrit, 32 percent; hemoglobin, 11 gldl; platelet count, 77,OOOIcu mm; prothrombin time, 65 percent; activated partial thromboplastin time (APnj 40 s (control, 36 s); fibrinogen, 685 mgldl; and 6brin(ogen) degradation products above 4,000 nglml; total billirubin was 3.4 mgldl with a direct component of 2.3 mgldl; creatine phosphokinase, 1,064 UIL with a MB fraction of27 UIL; serum alanine aminotransferase (ALl') was 75 UIL and serum aspartate transaminase (Am was 55 UIL. A pulmonary artery catheter was inserted, revealing a cardiac output of 6.3 Umin, pulmonary arterial pressure of 41/22 (31) mm Hg, and a pulmonary capillary wedge pressure of 10 mm Hg. Pulmonary and wedge position were con6rmed using published criteria consisting in analysis ofhemodynamic pressures, waveforms, and blood gas values. With the diagnosis ofadult respiratory distress syndrome (ARDS) probably secondary to fat embolism, we took nonheparinized samples of microvascular blood while the pulmonary artery catheter was in the wedge position. We rejected the first 10 ml of blood assuming that this amount represents "dead space" coming from terminal pulmonary arteries. These samples were immediately transferred to the pathology laboratory where they were centrifuged to separate the cellular and noncellular components of the 8uid. After this we 6xed the cells using 96 percent alcohol or letting them dry at room temperature. Preparations were stained using the Papanicolaou method in the first case or May-Grunwald-Giemsa and Sudan black method otherwise. Microscopic analysis of these samples with the May-GrunwaldGiemsa stain showed large numbers of globules (presumably fat) in a background of hemolyzed blood (Fig 1 and 2). These globules showed a positive pigmentation with the lipid-specific stain Sudan black supporting the initial assumption offat content in the globules (Fig 3). The patient was treated using mechanical ventilation with .:~
~
.'
,"(.l....~,
,. .' .'\41~~ .. .
..
.. '.
~.......
,,"
'\', .....
.•.." :.:
..
'
~ ~:.
""'. -.; . ""-:-".~' .~. > ......".' <.. ~.'~ ...... -.;. ~·r~"' . '·'\":' ·~A.·.~.· _., -,~. . ' ..... '¥....~.:-.\-~,. ~ -r~, ~ .,.'If.1II
. \... ~ .., ._ ..• ~ .._ - "V"; """., , ~ .... ~. • . _ .. ' ~iI' ...trIId...., .•.. ~.. . , ' • ., *':
.:c
.. I.
~""": '_.'.'-,.
~.........
'~".
...::
..
'or
~'. "
....
..
,".
·
~
...
!). III• • #",_,.:..-
.: • • • . . .
.....
•
III
•
.. ........ ...-
•..'
.."
. ,....
,....
_
'~~., •
.....
.'
4
,,"
"
WO'
.
~".
.
.•. ...-
"-..
:'.,~
' •• -
..
~~, .•'"
'i:I... ,.,. ,..
· .' l.·~·. :', .....~ t .... t;~ .. I ' • .•• ......... • .-:J ~: ~''P'T
• •",:,
""~' _ ..-' "'\'."". '.
. ~ ._ "'~~' ,., ~,.' ~.~ ..;.. ~~"'7 .,'d '(" .':'
,.. '''-..\ . ' - "'".' "'.".. ;" ....• '~." •. ,"t'.:.. ;' ~'~"\." •
,
.t
~1"""
,0
~
-,
•
• •11
•
\.~ •
•
'~
.•. .~ -~
. .t
...' ~ '* ;;~;1. .,~ t .....'-* ~., ., .' ~~z·", ""'S ~" . .f ......... .,..~....,. .... ~ ., :.'-"'~ T ..... " . ". J-.' ) " . 't . 1~:t'" • '
'1" _
....
.. ~
'..... . .~
•. ~
'.'
'
#f'
•
t
.:...... OIl.
.(
,.
;...
....
'.
:-,:-:-", ..... •
FIGURE 1. Pulmonary capillary sample from the patient showing globules (presumably fat) from the blood clot (May-GrunwaldGiemsa stain, original magnification X 40). Fat EmboIlsm Syndrome (castella st eI)
Swan-Ganz catheter. Despite this suggestive accuracy, we are unaware of other reports using pulmonary microvascular cytology in this syndrome. Our case corroborates the usefulness of a technique that we think is as specific as lung biopsy. From our experience, we would suggest that treatment of patients with suspected fat embolism should routinely include pulmonary artery catheterization and capillary blood sampling. REFERENCES
FIGURE 2. Closer view of Figure 2 focused on the globules (MayGrunwald-Giemsa stain, magnification x 200). positive end-expiratory pressure (PEEP) and methylprednisolone was added during 24 h. The patient improved and she could be weaned from mechanical ventilation on the sixth day. She was discharged from the ICU on the tenth day. DISCUSSION
Pulmonary microvascular cytology is a new technique that samples capillary lung blood obtained from a wedged SwanGanz catheter." Its usefulness has been shown in diagnosing a few cases of amniotic fluid embolism.'1-l3 Despite some discussion," Masson et al"," proposed capillary blood sampling as a useful diagnostic procedure in lymphangitic carcinomatosis of the lung and established it showed malignant cells in seven of eight samples from patients with Iymphangitic spread of carcinoma in the lungs. '6 Fat embolism is a protean syndrome elusive to diagnosis. 17 Classically it is diagnosed on a clinical basis alone. ,. Schonfeld et al'· have proposed an objective scoring system, but it cannot be used in patients with severe cerebral, thoracic, or abdominal injuries. Detection of fat globules in urine and blood measurement of serum lipase activity have been suggested to be useful in some patients, but these tests have never been evaluated critically. Recently, Chastre et al" have shown that BAL in patients with suspected fat embolism syndrome could be an accurate diagnostic procedure. In six of seven patients with suspected fat embolism, Masson and Ruggieri" could find fat globules in blood obtained from a wedged
FIGURE 3. The globules of Figures 1 and 2 showed a positive pigmentation with the lipid-Specific stain Sudan black supporting the initial assumption of fat content (Sudan black stain, original magnification x 1(0).
1 Evarts C. The fat embolism syndrome: a review. Surg Clin North Am 1970; 50:493-507 2 Dines D, Linscheid R, Didier E. Fat embolism syndrome. Mayo Clin Proc 1972; 47:237-40 3 Jacobson D, Terrence C, Reinmuth 0. The neurologic manifestations offat embolism. Neurology 1986; 36:847-51 4 Keith R, Mahoney L, Garvey M. Disseminated intravascular coagulation: an important feature of the fat embolism syndrome. Can Med Assoc J 1971; 105:74-6 5 Armstrong H, Kuenzig M, Peltier L. Lung lipase levels in normal rats with experimentally produced fat embolism. Proc Soc Exp BioI Med 1967; 124:959-61 6 Renne J, Wuthier R, House E, Cancro J, Hoaglund F. Fat macroglobulimia caused by fractures or total hip replacement. J Bone Joint Surg 1978; 6OA:613-17 7 Lepisto P, Alho A. Diagnostic features offat embolism syndrome. Acta Chir Scand 1975; 141:245-50 8 Chastre J, Fagon J, Soler P, Fichelle A, Dombret M, Huten D, et aI. Bronchoalveolar lavage for rapid diagnosis of the fat embolism syndrome in trauma patients. Ann Intern Med 1990; 113:583-88 9 Dorr L, Merkel C, Mellman M, Klein I. Fat emboli in bilateral total knee arthroplasty: predictive factors for neurologic manifestations. Clin Orthop 1989; 248:112-18 10 Byrick R, Kay J, Mullen J. Capnography is not as sensitive as pulmonary artery pressure monitoring in detecting marrow microembolism. Anesth Analg 1989; 68:94-100 11 Masson R, Ruggieri J. Pulmonary microvascular cytology: a new diagnostic application of the pulmonary artery catheter. Chest 1985; 88:908-14 12 Agia G, Wyatt L, Re E. Pulmonary microvascular cytology: a new diagnostic application of the pulmonary artery catheter. Chest 1986; 90:627-28 13 Vanmaele L, Noppen M, Vincken \v, De Catte L, Huyghens L. Transient left heart failure in amniotic Ouid embolism. Intensive Care Med 1990; 16:269-71 14 Spapen H, Umbrain V. Huyghens L. Case 30-1987: cytology of pulmonary blood samples. N Engl J Med 1988; 318:324 15 Masson R, Lukl P, Krilorian J, Evans G. Case 30-1987: cytology of pulmonary blood samples. N Engl J Med 1988; 318:324-25 16 Masson R, Krikorian J, Lukl P, Evans G, McGrath J. Pulmonary microvascular cytology in the diagnosis of Iymphangitic carcinomatosis. N Engl J Med 1989; 321:71-6 17 Fabian T, Hoots A, Stanford D, Patterson C, Mangiante E. Fat embolism syndrome: prospective evaluation in 92 fracture patients. Crit Care Med 1990; 18:42-6 18 Peltier L. Fat embolism: a perspective. Clin Orthop 1988; 232:263-70 19 Schonfeld S, Ploysonsang Y, DiLisio R, Crissman J, Miller E, Hammerschmidt D, et aI. Fat embolism prophylaxis with corticosteroids: a prospective study in high risk patients. Ann Intern Med 1983; 99:438-43 CHEST I 101 I 6 I JUNE. 1992
1711