A case of fatal fat embolism syndrome following cardiopulmonary bypass A case of fat embolism syndrome following median sternotomy and cardiopulmonary bypass, with fatal outcome, is described. This syndrome is present subclinically in the majority of these surgical procedures. The etiology and possible mechanisms of production are discussed; management is empirical and aimed at maintaining the function of embolized microvascular endothelium. A. J. Hodge, R. B. Dymock, and H. D. Sutherland, Adelaide, South Australia X. he presence of intravascular fat in pulmonary and systemic vessels following open-heart surgery is well documented, 1-6 but the fat is seldom present in quan- tities sufficient to cause symptoms or permanent sequelae. The mechanisms of this process and the origin of the lipid material are unknown. A case is presented to emphasize the fact that a technically satisfactory procedure may result in an irreversible process leading to death. Case report A. N., a 58-year-old European man, suffered an inferior myocardial infarction 7 years prior to admission. From that time he had experienced chest pain on exertion, which was successfully managed by alteration in life style. Five months prior to admission, the chest pain increased in frequency and severity; it was temporarily relieved by sublingual nitrates. The institution of beta blockade precipitated acute bron- chospasm, which was eventually relieved by long-term use of oral prednisolone, terbutaline (orally and by inhalation), and beclomethasone by inhalation. The chest pain was further controlled by oral Verapamil until 2 days before admission, when he experienced slow onset of central, nonradiating chest pain associated with sweating and weakness of both arms. On examination, pulse rate was regular at 72 beats per minute and blood pressure was 140/90 mm. Hg with normal cardiac findings. The right femoral pulse was diminished and the right dorsalis pedis was absent. The remaining pulses were normal. The lungs were clear. From the Cardiothoracic Surgical Unit, Royal Adelaide Hospital, Adelaide, and the Department of Pathology, the Queen Elizabeth Hospital, Adelaide, South Australia. Received for publication Feb. 3, 1976. Accepted for publication May 3, 1976. Address for reprints: Mr. H. D. Sutherland, Director, Cardiothoracic Surgical Unit, Royal Adelaide Hospital, North Terrace, Ade- laide, South Australia, 5000. The electrocardiogram indicated lateral ischemia, but con- centrations of serum lactic dehydrogenase and aspartic trans- aminase were within normal limits. The serum cholesterol level was 335 mg. per 100 ml. (normal 140 to 285 mg. per 100 ml.). A diagnosis of myocardial ischemia without infarc- tion was made. Coronary arteriograms revealed a subtotal obstruction of the left anterior descending artery proximally and a total obstruction of the right coronary artery at its origin. The left main artery and the circumflex system were normal. Left ventriculogram indicated poor inferior movement with mod- erate hypokinesis anteriorly. Two months following arteriographic study, reversed au- togenous saphenous vein bypass grafts were inserted into the right and left anterior descending coronary arteries. The operation was performed by use of a Melrose disc oxygenator with whole blood prime and moderate hypothermia. Total bypass time was 113 minutes. The sternum was divided with an electric saw and secured during closure with peristernal and transsternal wires with the use of a tightening device according to our standard practice. Wax was used to achieve sternal hemostasis. The sternal trabeculae were not noted to be unusually soft. Early the following morning the patient was returned to the operating theater for evacuation of retrosternal and epigastric clot, the appearances suggesting that bleeding had occurred either from the back of the sternum or from the incision in general. There was no evidence elsewhere at this time of a bleeding tendency. Extubation was possible by 24 hours and, although con- scious his level of consciousness was within normal limits, his coughing ability was poor. Over the succeeding 24 hours, the patient became confused, respiratory effort declined, and reintubation was necessary (arterial Po 2 62 and Pco 2 56 mm. Hg while breathing 50 per cent oxygen). Cardiovascular state remained satisfactory throughout. Fifty-four hours after the commencement of the first operation and 48 hours after the second, a punctate hemor- rhagic rash was noted over both flanks and in the conjunctivae (Fig. 1). Increased ventilation was necessary to achieve a normal arterial Pco 2 , intravenous furosemide was given to induce dehydration, and postoperative hydrocortisone (to 202
A case of fatal fat embolism syndrome following cardiopulmonary bypass
Jan 30, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A case of fatal fat embolism syndrome following cardiopulmonary bypassA case of fatal fat embolism syndrome following cardiopulmonary bypass A case of fat embolism syndrome following median sternotomy and cardiopulmonary bypass, with fatal outcome, is described. This syndrome is present subclinically in the majority of these surgical procedures. The etiology and possible mechanisms of production are discussed; management is empirical and aimed at maintaining the function of embolized microvascular endothelium.
A. J. Hodge, R. B. Dymock, and H. D. Sutherland, Adelaide, South Australia
X. he presence of intravascular fat in pulmonary and systemic vessels following open-heart surgery is well documented,1-6 but the fat is seldom present in quan- tities sufficient to cause symptoms or permanent sequelae. The mechanisms of this process and the origin of the lipid material are unknown.
A case is presented to emphasize the fact that a technically satisfactory procedure may result in an irreversible process leading to death.
Case report
A. N., a 58-year-old European man, suffered an inferior myocardial infarction 7 years prior to admission. From that time he had experienced chest pain on exertion, which was successfully managed by alteration in life style. Five months prior to admission, the chest pain increased in frequency and severity; it was temporarily relieved by sublingual nitrates.
The institution of beta blockade precipitated acute bron- chospasm, which was eventually relieved by long-term use of oral prednisolone, terbutaline (orally and by inhalation), and beclomethasone by inhalation.
The chest pain was further controlled by oral Verapamil until 2 days before admission, when he experienced slow onset of central, nonradiating chest pain associated with sweating and weakness of both arms.
On examination, pulse rate was regular at 72 beats per minute and blood pressure was 140/90 mm. Hg with normal cardiac findings. The right femoral pulse was diminished and the right dorsalis pedis was absent. The remaining pulses were normal. The lungs were clear.
From the Cardiothoracic Surgical Unit, Royal Adelaide Hospital, Adelaide, and the Department of Pathology, the Queen Elizabeth Hospital, Adelaide, South Australia.
Received for publication Feb. 3, 1976. Accepted for publication May 3, 1976. Address for reprints: Mr. H. D. Sutherland, Director, Cardiothoracic
Surgical Unit, Royal Adelaide Hospital, North Terrace, Ade- laide, South Australia, 5000.
The electrocardiogram indicated lateral ischemia, but con- centrations of serum lactic dehydrogenase and aspartic trans- aminase were within normal limits. The serum cholesterol level was 335 mg. per 100 ml. (normal 140 to 285 mg. per 100 ml.). A diagnosis of myocardial ischemia without infarc- tion was made.
Coronary arteriograms revealed a subtotal obstruction of the left anterior descending artery proximally and a total obstruction of the right coronary artery at its origin. The left main artery and the circumflex system were normal. Left ventriculogram indicated poor inferior movement with mod- erate hypokinesis anteriorly.
Two months following arteriographic study, reversed au- togenous saphenous vein bypass grafts were inserted into the right and left anterior descending coronary arteries. The operation was performed by use of a Melrose disc oxygenator with whole blood prime and moderate hypothermia. Total bypass time was 113 minutes. The sternum was divided with an electric saw and secured during closure with peristernal and transsternal wires with the use of a tightening device according to our standard practice. Wax was used to achieve sternal hemostasis. The sternal trabeculae were not noted to be unusually soft.
Early the following morning the patient was returned to the operating theater for evacuation of retrosternal and epigastric clot, the appearances suggesting that bleeding had occurred either from the back of the sternum or from the incision in general. There was no evidence elsewhere at this time of a bleeding tendency.
Extubation was possible by 24 hours and, although con- scious his level of consciousness was within normal limits, his coughing ability was poor. Over the succeeding 24 hours, the patient became confused, respiratory effort declined, and reintubation was necessary (arterial Po2 62 and Pco2 56 mm. Hg while breathing 50 per cent oxygen). Cardiovascular state remained satisfactory throughout.
Fifty-four hours after the commencement of the first operation and 48 hours after the second, a punctate hemor- rhagic rash was noted over both flanks and in the conjunctivae (Fig. 1). Increased ventilation was necessary to achieve a normal arterial Pco2, intravenous furosemide was given to induce dehydration, and postoperative hydrocortisone (to
202
Fig. 1. Subconjunctival hemorrhages.
Fig. 2. Cerebral hemispheres—hemorrhages in white matter.
cover the adrenal suppression present preoperatively) was supplemented by dexamethasone intravenously. On the eve- ning of the second postoperative day sinus rhythm suddenly gave way to ventricular fibrillation which could not be reversed.
At postmortem examination, numerous petechial hemor- rhages were apparent over the thorax, abdomen, upper and lower limbs, in the subconjunctivae (Fig. 1), and in the visceral pleura and pericardium.
The brain weight was normal but the cut surface revealed large numbers of petechiae within the white matter of the cerebral hemispheres (Fig. 2), cerebellum, midbrain, and hindbrain. In addition to routine hematoxylin and eosin sections, blocks of cerebral and lung tissue were submitted to immediate frozen section with Sudan staining (oil red O), and formalin-fixed blocks of brain, lung, and kidney were submit- ted to postfixation with osmium tetroxide.
With hematoxylin and eosin staining, clear spaces were apparent in the central zones of petechial hemorrhages (Fig. 3) and within glomerular capillaries (Fig. 4). The Sudan and
Fatal fat embolism after cardiopulmonary bypass 2 0 3
Fig. 3. Brain—clear spaces of dissolved lipid with surround- ing hemorrhage. (Hematoxylin and eosin, X120.)
Fig. 4. Clear spaces of dissolved lipid within glomerular capillary loops. (Hematoxylin and eosin, X300.)
osmium tetroxide stains revealed that these clear spaces contained lipid material, confirming the clinical diagnosis of systemic fat embolism syndrome.
There was gross atherosclerosis of the left and right coronary artieries; both saphenous vein grafts were widely patent. The liver did not contain abnormal fat deposits.
Discussion
This is the first recognized instance of the patient dying of the fat embolism syndrome in 2,000 median sternotomies with bypass performed in this Unit. The syndrome occurrs in many other states such as external cardiac massage,7 bony trauma,8 high-altitude flights, postpartum state, alcoholism, poisonings,9 diabetes, severe burns, severe infections, inhalation anesthesia, metabolic disorders, neoplasia, osteomyelitis, decom- pression sickness, steroid-induced fatty liver, and sickle cell disease.10 Clearly, a theory to encompass
2 0 4 Hodge, Dymock, Sutherland The Journal of
Thoracic and Cardiovascular Surgery
such a variety of states must take into account multiple etiologies to arrive at an acceptable pathogenesis.
The simplest hypothesis is the intravasation or me- chanical process,8,'' which has been proposed to explain fat embolism following bone fractures and relies on unique properties of bone favorable to embolization— high fat content, high vascularity, rigidity, and pres- ence of thin-walled veins held open by the bony trabeculae. Liquid fat is forced into the veins by the pressure of inflammatory edema; hemopoietic tissue has been described in lung as an accompaniment of fat embolism.7, 12
The second hypothesis held to explain fat embolism is the physicochemical theory,13, 14 which suggests that, with trauma, catecholamines mobilize free fatty acids. This in turn leads to the synthesis of very low-density lipoproteins which are cholesterol transport vehicles. These lipoproteins "attract" cholesterol from the bloodstream to such a degree that large numbers of chylomicra coalesce to form droplet emboli. This explains why fat emboli contain 10 to 30 per cent cholesterol and cholesterol esters, when depot or mar- row fat contains less than 1 per cent cholesterol.
Another proposal is that the blood-gas interface existing in film and bubble oxygenators upsets the balance of surface tension or electrical forces that maintain fatty material in an emulsified state, allowing it to coalesce into droplets.
All of these mechanisms have been shown to be present in a procedure utilizing sternal division with cardiopulmonary bypass: Right atrial fat globules ap- pear in the second and third minutes following sternal division,9 cardiopulmonary bypass regularly causes systemic fat emboli to all organs,4 and the production of fat globules has been used as a measure of the efficiency of various oxygenators.5 It may also be possible for sternal fat to enter the circulation with the pressure of the local inflammatory reaction following sternal wire fixation.
Aspiration of fat into the bypass machine from the pericardial cavity occurs regularly, but the use of 15 to 40 /Li filters in the cardiotomy or arterial line has been shown to be ineffective in reducing neurologic or pulmonary complications,15 although this has not been the experience of many units, including our own.
Therefore, although intravascular fat will appear in the majority of patients undergoing cardiopulmonary bypass, the fat embolism syndrome will not be diag- nosed in life unless there is sufficient globular fat obstructing and damaging the microcirculation to cause the clinical syndrome of petechial rash, pyrexia, vari- ous neurologic syndromes, tachycardia, hypotension,
and falling arterial Po2 with dyspnea, tachypnea, and cyanosis. Searching for fat in the sputum, urine, and retinal vessels is not contributory.15 There will be many patients not exhibiting the full syndrome who will be labeled as having "the postperfusion syndrome" or "pump lung" or who will exhibit transient neurologic phenomena for which there is no satisfactory explana- tion.
Recognition of patients at risk and prevention of significant fat embolism is not possible at present.
Patients who have a high cholesterol level before operation have a smaller proportionate rise in lipid levels after operation than those with low preoperative cholesterol levels.15 There are no equivalent data for diabetes or other states of raised serum lipids.
Management of the established syndrome is empiri- cal, as pathogenesis is unknown. The use of heparin in full anticoagulant or in subanticoagulant doses has been tried, with the rationale of stimulating serum lipopro- tein lipase to split free fatty acids from the cholesterol molecule,16 the remaining partial glycerides being more soluble than cholesterol.17
However, as mentioned previously, release of free fatty acids may perpetuate the syndrome, and these substances are toxic to the microvascular en- dothelium.10 The role of heparin used in full an- ticoagulant doses for cardiopulmonary bypass is speculative.
The present regimen of management in the general Intensive Care Unit at the Royal Adelaide Hospital18 is directed primarily at the pulmonary manifestations. The tendency toward interstitial and intra-alveolar edema caused by endothelial damage is offset by lowering left atrial pressure, by using fluid restriction and diuretics (furosemide in large doses intravenously), and by maintaining arterial Po2 within normal limits— usually requiring intermittent positive-pressure breath- ing with 50 to 100 per cent inspired oxygen, Intrave- nous steroids have been used recently, but without demonstrable benefit, as the damage to vascular en- dothelium is well advanced before definitive diagnosis is made.
R E F E R E N C E S 1 Ellison, L. T., McPherson, J. C , Jr., Anabtawi, I. N.,
and Ellison, R. G.: Incidence of Free Fat in the Lung During Open-Heart Surgery, Ann. Thorac. Surg. 7: 509, 1969.
2 Hill, J. D., Aguilar, M. J., Baranco, A., de Lanerolle, P., and Gerbode, F.: Neuropathological Manifestations of Cardiac Surgery, Ann. Thorac. Surg. 7: 409, 1969.
3 Price, D. L., and Harris, J.: Cholesterol Emboli in Cerebral Arteries as a Complication of Retrograde Aortic
Volume 72 Number 2 August, 1976
Fatal fat embolism after cardiopulmonary bypass 2 0 5
Perfusion During Cardiac Surgery, Neurology (Min- neap.) 20: 1209, 1970.
4 Miller, J. A., Fonkalsrud, E. W., Latta, H. L., and Maloney, I. V., Jr.: Fat Embolism Associated with Extracorporeal Circulation and Blood Transfusion, Surgery 51: 448, 1962.
5 Owens, G., Adams, J. E., and Scott, H. W.: Embolic Fat as a Measure of Adequacy of Various Oxygenators, J. Appl. Physiol. 15: 999, 1960.
6 Wright, E. S., Sarkozy, E., Dobell, A. R. C , and Murphy, D. R.: Fat Globulemia in Extracorporeal Circu- lation, Surgery 53: 500, 1963.
7 Lane, J. H., and Merkel, W. C : External Cardiac Massage: A Cause of Bone Marrow and Fat Emboli, South. Med. J. 58:450, 1965.
8 Gauss, H.: Pathology of Fat Embolism, Arch. Surg. 9: 593, 1924.
9 Wukasch, D. C . Malloy, K. P., Rubio, P. A., Reed, C. C , Sandiford. F. M., Reul, G. J., Milam, J. D., and Cooley, D. C : Fat Embolization Resulting From Median Sternotomy, Tex. Med. 71: 35, 1975.
10 Miller, G. A.: Fat Embolism: A Comprehensive Review, J. Oral Surg. 33: 91, 1975.
11 Whitenack, S. H., and Hausberger, F. X.: Intravazation of Fat From the Bone Marrow Cavity, Am. J. Pathol. 65: 335, 1971.
12 Yanoff, M.: Incidence of Bone Marrow Embolism due to Closed Chest Cardiac Massage, N. Engl. J. Med. 269: 837, 1963.
13 Lehman, E. P., and Moore, R. M.: Fat Embolism: Including Experimental Production Without Trauma, Arch. Surg. 14: 621, 1927.
14 Hillman, J. W., and Le Quire, V. S.: Lipid Metabolism and Fat Embolism After Trauma: The Contribution of Serum Lipoproteins to Embolic Fat, Surg. Forum 19: 465, 1968.
15 Arrants, J. E., Gadsden, R. H., Huggins, M. B., and Lee, W. H.: Effects of Extracorporeal Circulation Upon Blood Lipids, Ann. Thorac. Surg. 15: 230, 1973.
16 Cobb, C. A., Le Quire, V.S.,Gray,M. E., and Hillman, J. W.: Therapy of Traumatic Fat Embolism With In- travenous Fluids and Heparin, Surg. Forum 9: 751, 1958.
17 Meng, H. C , Hollett, C , and Cole, W. E.: Some Factors Affecting the Clearing of Neutral Fat Embolism by Post-heparin Plasma, Am. J. Physiol. 179: 314, 1954.
18 Worthley, T. E.: Personal communication, 1976.
A. J. Hodge, R. B. Dymock, and H. D. Sutherland, Adelaide, South Australia
X. he presence of intravascular fat in pulmonary and systemic vessels following open-heart surgery is well documented,1-6 but the fat is seldom present in quan- tities sufficient to cause symptoms or permanent sequelae. The mechanisms of this process and the origin of the lipid material are unknown.
A case is presented to emphasize the fact that a technically satisfactory procedure may result in an irreversible process leading to death.
Case report
A. N., a 58-year-old European man, suffered an inferior myocardial infarction 7 years prior to admission. From that time he had experienced chest pain on exertion, which was successfully managed by alteration in life style. Five months prior to admission, the chest pain increased in frequency and severity; it was temporarily relieved by sublingual nitrates.
The institution of beta blockade precipitated acute bron- chospasm, which was eventually relieved by long-term use of oral prednisolone, terbutaline (orally and by inhalation), and beclomethasone by inhalation.
The chest pain was further controlled by oral Verapamil until 2 days before admission, when he experienced slow onset of central, nonradiating chest pain associated with sweating and weakness of both arms.
On examination, pulse rate was regular at 72 beats per minute and blood pressure was 140/90 mm. Hg with normal cardiac findings. The right femoral pulse was diminished and the right dorsalis pedis was absent. The remaining pulses were normal. The lungs were clear.
From the Cardiothoracic Surgical Unit, Royal Adelaide Hospital, Adelaide, and the Department of Pathology, the Queen Elizabeth Hospital, Adelaide, South Australia.
Received for publication Feb. 3, 1976. Accepted for publication May 3, 1976. Address for reprints: Mr. H. D. Sutherland, Director, Cardiothoracic
Surgical Unit, Royal Adelaide Hospital, North Terrace, Ade- laide, South Australia, 5000.
The electrocardiogram indicated lateral ischemia, but con- centrations of serum lactic dehydrogenase and aspartic trans- aminase were within normal limits. The serum cholesterol level was 335 mg. per 100 ml. (normal 140 to 285 mg. per 100 ml.). A diagnosis of myocardial ischemia without infarc- tion was made.
Coronary arteriograms revealed a subtotal obstruction of the left anterior descending artery proximally and a total obstruction of the right coronary artery at its origin. The left main artery and the circumflex system were normal. Left ventriculogram indicated poor inferior movement with mod- erate hypokinesis anteriorly.
Two months following arteriographic study, reversed au- togenous saphenous vein bypass grafts were inserted into the right and left anterior descending coronary arteries. The operation was performed by use of a Melrose disc oxygenator with whole blood prime and moderate hypothermia. Total bypass time was 113 minutes. The sternum was divided with an electric saw and secured during closure with peristernal and transsternal wires with the use of a tightening device according to our standard practice. Wax was used to achieve sternal hemostasis. The sternal trabeculae were not noted to be unusually soft.
Early the following morning the patient was returned to the operating theater for evacuation of retrosternal and epigastric clot, the appearances suggesting that bleeding had occurred either from the back of the sternum or from the incision in general. There was no evidence elsewhere at this time of a bleeding tendency.
Extubation was possible by 24 hours and, although con- scious his level of consciousness was within normal limits, his coughing ability was poor. Over the succeeding 24 hours, the patient became confused, respiratory effort declined, and reintubation was necessary (arterial Po2 62 and Pco2 56 mm. Hg while breathing 50 per cent oxygen). Cardiovascular state remained satisfactory throughout.
Fifty-four hours after the commencement of the first operation and 48 hours after the second, a punctate hemor- rhagic rash was noted over both flanks and in the conjunctivae (Fig. 1). Increased ventilation was necessary to achieve a normal arterial Pco2, intravenous furosemide was given to induce dehydration, and postoperative hydrocortisone (to
202
Fig. 1. Subconjunctival hemorrhages.
Fig. 2. Cerebral hemispheres—hemorrhages in white matter.
cover the adrenal suppression present preoperatively) was supplemented by dexamethasone intravenously. On the eve- ning of the second postoperative day sinus rhythm suddenly gave way to ventricular fibrillation which could not be reversed.
At postmortem examination, numerous petechial hemor- rhages were apparent over the thorax, abdomen, upper and lower limbs, in the subconjunctivae (Fig. 1), and in the visceral pleura and pericardium.
The brain weight was normal but the cut surface revealed large numbers of petechiae within the white matter of the cerebral hemispheres (Fig. 2), cerebellum, midbrain, and hindbrain. In addition to routine hematoxylin and eosin sections, blocks of cerebral and lung tissue were submitted to immediate frozen section with Sudan staining (oil red O), and formalin-fixed blocks of brain, lung, and kidney were submit- ted to postfixation with osmium tetroxide.
With hematoxylin and eosin staining, clear spaces were apparent in the central zones of petechial hemorrhages (Fig. 3) and within glomerular capillaries (Fig. 4). The Sudan and
Fatal fat embolism after cardiopulmonary bypass 2 0 3
Fig. 3. Brain—clear spaces of dissolved lipid with surround- ing hemorrhage. (Hematoxylin and eosin, X120.)
Fig. 4. Clear spaces of dissolved lipid within glomerular capillary loops. (Hematoxylin and eosin, X300.)
osmium tetroxide stains revealed that these clear spaces contained lipid material, confirming the clinical diagnosis of systemic fat embolism syndrome.
There was gross atherosclerosis of the left and right coronary artieries; both saphenous vein grafts were widely patent. The liver did not contain abnormal fat deposits.
Discussion
This is the first recognized instance of the patient dying of the fat embolism syndrome in 2,000 median sternotomies with bypass performed in this Unit. The syndrome occurrs in many other states such as external cardiac massage,7 bony trauma,8 high-altitude flights, postpartum state, alcoholism, poisonings,9 diabetes, severe burns, severe infections, inhalation anesthesia, metabolic disorders, neoplasia, osteomyelitis, decom- pression sickness, steroid-induced fatty liver, and sickle cell disease.10 Clearly, a theory to encompass
2 0 4 Hodge, Dymock, Sutherland The Journal of
Thoracic and Cardiovascular Surgery
such a variety of states must take into account multiple etiologies to arrive at an acceptable pathogenesis.
The simplest hypothesis is the intravasation or me- chanical process,8,'' which has been proposed to explain fat embolism following bone fractures and relies on unique properties of bone favorable to embolization— high fat content, high vascularity, rigidity, and pres- ence of thin-walled veins held open by the bony trabeculae. Liquid fat is forced into the veins by the pressure of inflammatory edema; hemopoietic tissue has been described in lung as an accompaniment of fat embolism.7, 12
The second hypothesis held to explain fat embolism is the physicochemical theory,13, 14 which suggests that, with trauma, catecholamines mobilize free fatty acids. This in turn leads to the synthesis of very low-density lipoproteins which are cholesterol transport vehicles. These lipoproteins "attract" cholesterol from the bloodstream to such a degree that large numbers of chylomicra coalesce to form droplet emboli. This explains why fat emboli contain 10 to 30 per cent cholesterol and cholesterol esters, when depot or mar- row fat contains less than 1 per cent cholesterol.
Another proposal is that the blood-gas interface existing in film and bubble oxygenators upsets the balance of surface tension or electrical forces that maintain fatty material in an emulsified state, allowing it to coalesce into droplets.
All of these mechanisms have been shown to be present in a procedure utilizing sternal division with cardiopulmonary bypass: Right atrial fat globules ap- pear in the second and third minutes following sternal division,9 cardiopulmonary bypass regularly causes systemic fat emboli to all organs,4 and the production of fat globules has been used as a measure of the efficiency of various oxygenators.5 It may also be possible for sternal fat to enter the circulation with the pressure of the local inflammatory reaction following sternal wire fixation.
Aspiration of fat into the bypass machine from the pericardial cavity occurs regularly, but the use of 15 to 40 /Li filters in the cardiotomy or arterial line has been shown to be ineffective in reducing neurologic or pulmonary complications,15 although this has not been the experience of many units, including our own.
Therefore, although intravascular fat will appear in the majority of patients undergoing cardiopulmonary bypass, the fat embolism syndrome will not be diag- nosed in life unless there is sufficient globular fat obstructing and damaging the microcirculation to cause the clinical syndrome of petechial rash, pyrexia, vari- ous neurologic syndromes, tachycardia, hypotension,
and falling arterial Po2 with dyspnea, tachypnea, and cyanosis. Searching for fat in the sputum, urine, and retinal vessels is not contributory.15 There will be many patients not exhibiting the full syndrome who will be labeled as having "the postperfusion syndrome" or "pump lung" or who will exhibit transient neurologic phenomena for which there is no satisfactory explana- tion.
Recognition of patients at risk and prevention of significant fat embolism is not possible at present.
Patients who have a high cholesterol level before operation have a smaller proportionate rise in lipid levels after operation than those with low preoperative cholesterol levels.15 There are no equivalent data for diabetes or other states of raised serum lipids.
Management of the established syndrome is empiri- cal, as pathogenesis is unknown. The use of heparin in full anticoagulant or in subanticoagulant doses has been tried, with the rationale of stimulating serum lipopro- tein lipase to split free fatty acids from the cholesterol molecule,16 the remaining partial glycerides being more soluble than cholesterol.17
However, as mentioned previously, release of free fatty acids may perpetuate the syndrome, and these substances are toxic to the microvascular en- dothelium.10 The role of heparin used in full an- ticoagulant doses for cardiopulmonary bypass is speculative.
The present regimen of management in the general Intensive Care Unit at the Royal Adelaide Hospital18 is directed primarily at the pulmonary manifestations. The tendency toward interstitial and intra-alveolar edema caused by endothelial damage is offset by lowering left atrial pressure, by using fluid restriction and diuretics (furosemide in large doses intravenously), and by maintaining arterial Po2 within normal limits— usually requiring intermittent positive-pressure breath- ing with 50 to 100 per cent inspired oxygen, Intrave- nous steroids have been used recently, but without demonstrable benefit, as the damage to vascular en- dothelium is well advanced before definitive diagnosis is made.
R E F E R E N C E S 1 Ellison, L. T., McPherson, J. C , Jr., Anabtawi, I. N.,
and Ellison, R. G.: Incidence of Free Fat in the Lung During Open-Heart Surgery, Ann. Thorac. Surg. 7: 509, 1969.
2 Hill, J. D., Aguilar, M. J., Baranco, A., de Lanerolle, P., and Gerbode, F.: Neuropathological Manifestations of Cardiac Surgery, Ann. Thorac. Surg. 7: 409, 1969.
3 Price, D. L., and Harris, J.: Cholesterol Emboli in Cerebral Arteries as a Complication of Retrograde Aortic
Volume 72 Number 2 August, 1976
Fatal fat embolism after cardiopulmonary bypass 2 0 5
Perfusion During Cardiac Surgery, Neurology (Min- neap.) 20: 1209, 1970.
4 Miller, J. A., Fonkalsrud, E. W., Latta, H. L., and Maloney, I. V., Jr.: Fat Embolism Associated with Extracorporeal Circulation and Blood Transfusion, Surgery 51: 448, 1962.
5 Owens, G., Adams, J. E., and Scott, H. W.: Embolic Fat as a Measure of Adequacy of Various Oxygenators, J. Appl. Physiol. 15: 999, 1960.
6 Wright, E. S., Sarkozy, E., Dobell, A. R. C , and Murphy, D. R.: Fat Globulemia in Extracorporeal Circu- lation, Surgery 53: 500, 1963.
7 Lane, J. H., and Merkel, W. C : External Cardiac Massage: A Cause of Bone Marrow and Fat Emboli, South. Med. J. 58:450, 1965.
8 Gauss, H.: Pathology of Fat Embolism, Arch. Surg. 9: 593, 1924.
9 Wukasch, D. C . Malloy, K. P., Rubio, P. A., Reed, C. C , Sandiford. F. M., Reul, G. J., Milam, J. D., and Cooley, D. C : Fat Embolization Resulting From Median Sternotomy, Tex. Med. 71: 35, 1975.
10 Miller, G. A.: Fat Embolism: A Comprehensive Review, J. Oral Surg. 33: 91, 1975.
11 Whitenack, S. H., and Hausberger, F. X.: Intravazation of Fat From the Bone Marrow Cavity, Am. J. Pathol. 65: 335, 1971.
12 Yanoff, M.: Incidence of Bone Marrow Embolism due to Closed Chest Cardiac Massage, N. Engl. J. Med. 269: 837, 1963.
13 Lehman, E. P., and Moore, R. M.: Fat Embolism: Including Experimental Production Without Trauma, Arch. Surg. 14: 621, 1927.
14 Hillman, J. W., and Le Quire, V. S.: Lipid Metabolism and Fat Embolism After Trauma: The Contribution of Serum Lipoproteins to Embolic Fat, Surg. Forum 19: 465, 1968.
15 Arrants, J. E., Gadsden, R. H., Huggins, M. B., and Lee, W. H.: Effects of Extracorporeal Circulation Upon Blood Lipids, Ann. Thorac. Surg. 15: 230, 1973.
16 Cobb, C. A., Le Quire, V.S.,Gray,M. E., and Hillman, J. W.: Therapy of Traumatic Fat Embolism With In- travenous Fluids and Heparin, Surg. Forum 9: 751, 1958.
17 Meng, H. C , Hollett, C , and Cole, W. E.: Some Factors Affecting the Clearing of Neutral Fat Embolism by Post-heparin Plasma, Am. J. Physiol. 179: 314, 1954.
18 Worthley, T. E.: Personal communication, 1976.
Related Documents
