- Email: [email protected]
Air Embolism during Tunneled Central Catheter Placement Performed without General Anesthesia in Children: A Potentially Serious Complication
lnterventional Radiologist at Work
Air Embolism during Tunneled Central Catheter Placement Performed without General Anesthesia in Children: A Potentially Serious Complication1 Frank P. Morello, MD James 5. Donaldson, MD Martha C. Saker, MD Jackson T. Norman, MD
Index terms: Catheters and catheterization, complications * Central venous access Embolism, air
JVIR 1999; 10:781-784 Abbreviations: CPR = cardiopulmonary resuscitation, GA = general anesthesia, IV = intravenous, PPV = positive pressure ventilation
From the Children's Memorial Medical Center, Northwestern University Medical School, 2300 Children's Plaza, Chicago, IL 60614. Received September 9, 1998; revision requested October 19; revision received December 21; accepted December 28. Address correspondence to F.P.M. 0 SCVIR, 1999
CENTRAL venous catheters have had an increasingly important role in a variety of patient care situations, including long-term antibiotic therapy, chemotherapy, and nutritional support. The recent past has seen a gradual transition from placement of vascular access catheters by surgeons to placement by interventional radiologists (1).The interventional radiology service places a majority of the vascular access devices at our children's hospital, including peripherally inserted central catheters, tunneled central venous catheters, temporary and permanent hemodialysis catheters, and subcutaneous ports. Most procedures performed by our interventional radiology service in children can be successfully completed with use of intravenous (IV) sedation, and a few require general anesthesia (GA). Key advantages of GA over IV sedation include the ability to have positive pressure ventilation (PPV) or controlled apnea during the procedure. We report our experience of venous air embolism in three small children during placement of tunneled central venous catheters when GA was not used.
1 TECHNIQUE The guidelines written by the Committee on Drugs of the American Academy of Pediatrics were followed for all procedures requiring sedation (2,3). While sedated, all
patients had continuous pulse oximetry monitoring and vital signs including blood pressure, heart rate, and respiration rate were monitored and recorded every 5 minutes. The technique used for tunneled central catheter la cement in children was identical t o techniques described for adults (4.5). . , All venipunctures were done with use of ultrasound guidance and micropuncture access (Cook, Bloomington, IN) into the right internal jugular vein. As a modification to the standard technique, a subcutaneous tunnel was then formed between the catheter exit and the venipuncture sites using a blunt tip needle (Hawkins needle; Cook). A guide wire was passed through the needle and, after the needle was removed, a peel-away sheath was advanced through the tunnel to facilitate passage of the catheter. This peel-away sheath was then removed, leaving the catheter in the subcutaneous tunnel. The inner dilator and wire of the micropuncture set in the internal jugular vein were exchanged for a 0.035-inch guide wire, and the outer dilator was then exchanged for an appropriately sized peel-away sheath, which was inserted into the vein (Fig 1).The next step was the critical one, where the inner dilator and wire were removed from the peel-away sheath temporarily exposing the open sheath to open air (Fig 2). The peel-away sheath was immediately pinched after removing ,
782
Air Embolism during Tunneled Central Catheter Placement
June 1999 JVIR
Figures 1, 2. (1) Placement of a tunneled central venous catheter. After venipuncture of the right internal jugular vein, a
peel-away sheath with inner dilator is advanced over a guide wire into the superior vena cava. The central venous system is not open to air. The catheter has been advanced through the subcutaneous tunnel. (2) The inner dilator and wire have been removed so that the catheter may be advanced into the peel-away sheath. The central venous system is open to air. the inner dilator and wire, and the catheter was quickly inserted through the open sheath into the superior vena cava, with the sheath being pinched between each forward push of the catheter. After catheter insertion, the peel-away sheath was removed, the catheter was sutured, and a sterile dressing was applied.
I CASE REPORTS Case 1.-A 3-year-old boy with cerebral palsy and developmental delay had osteomyelitis of the left tibia. Placement of a central catheter was needed for long-term IV antibiotic administration. The patient was not sedated for the procedure because of thick copious orotracheal secretions and because the patient did not have purposeful movements. The right internal jugular vein was used for access and our described technique for tunneled central catheter placement was followed. During the critical step, which is the very short time when the 5-F sheath was open to atmospheric pressure, the patient began to cough. This was thought to be due to his secretions. A soft hissing sound through the sheath was heard and the patient became hy-
poxic (02saturation, 85%) and bradycardic (heart rate, 60). Supplemental oxygen was administered and the child promptly recovered to baseline vital signs. The 4.2-F (Bard, Salt Lake City, UT) catheter was successfully placed as the patient recovered. Case 2.-An 18-month-old boy with end-stage renal disease secondary to posterior urethral valves needed a tunneled hemodialysis catheter while awaiting a renal transplant. The patient was sedated with titrated doses of ketamine and midazolam but remained slightly agitated throughout the procedure. The patient had undergone previous interventional radiologic procedures and was quite difficult to sedate. He would remain agitated at dose levels that would be adequate for other children, but be hypotensive a t higher levels. After placement of a 10-F peel-away sheath into the internal jugular vein, a 9-F catheter (Medcomp, Harleysville, PA) was being advanced into the sheath when the patient began crying. A hissing sound was heard around the catheter, which was already in the sheath. This corresponded to a 10% drop in oxygen saturation. The patient recovered quickly and reu
mained stable with supplemental oxygen as the procedure was completed. Case 3.-A 10-month-old girl with congenital leukemia was referred for central catheter placement for chemotherapy and future bone marrow transplant. She was sedated with titrated doses of ketamine and midazolam. During the critical step, as the 7-F double-lumen catheter (Bard, Salt Lake City, UT) was about to be advanced through the open 8-F sheath, the patient became irritable and took a deep spontaneous breath. Air was heard being loudly sucked into the sheath. The patient immediately became hypotensive and profoundly hypoxic with ensuing cardiorespiratory arrest. After vigorous cardiopulmonary resuscitation (CPR), including intubation, the patient was stabilized. During initial CPR and intubation, the sheath was held tightly pinched by the radiologist. After the patient was intubated, chest compressions were briefly interrupted while the central catheter was advanced into the sheath, and the sheath was then removed. Resuscitative efforts continued, with the central catheter being used to administer drugs and fluids. The
Morello et a1
783
Volume 10 Number 6
patient had no sequelae from this episode.
I DISCUSSION In the three cases reported, venous air embolism was a complication of tunneled central venous catheter placement. All three incidents occurred under similar circumstances. That is, all patients were breathing spontaneously during a critical step of the catheter insertion. A centrally placed sheath was temporarily exposed to open air during exchange of a dilator for the catheter. None of the patients could hold their breath on command (two being deeply sedated, and one a developmentally delayed child who was not sedated). Spontaneous breaths generated sufficient negative pressure to cause venous air embolism during the critical catheter exchange. Venous air embolism in two of the children were minor, but a near disaster occurred in the third child. The effects of venous air embolism can vary from clinically insignificant to com~letecardiovascular collapse. Venous air embolism can be the most dramatic and, potentially, the most serious complication of central venous catheters, with a reported mortality of 29%-36% (6). The volume of air tolerated is unknown but accidental injections of air of between 100 mL and 300 mL have been reported to be fatal (7). This volume becomes relevant considering that it has been calculated that 100 mL of air per second can flow through a 14-gauge needle, with only a 5 cm H,O pressure gradient (8).A 14-gauge needle has an inner diameter slightly larger than a 6-F sheath. The sheath sizes used in our case reports were 5 F, 8 F, and 10 F. A 10-F sheath has a cross-sectional area 2.7 times larger than that of a 6-F sheath. One can easily see how a large air bolus can be entrained through these vascular sheaths. The factors that determine the morbidity and mortality of venous air embolism include the volume of
air and the rate at which air enters tion allows air in the pulmonary the venous system, as well as the outflow tract to migrate to the apex position of the patient at the time of the right ventricle and permits the venous air embolism occurred flow of blood to the lungs (15). Oth(9,lO). Intrinsically, the adaptability ers have recommended leaving the of the patient's cardiovascular syspatient supine, rather than turning tem is probably a factor as well. to a left lateral decubitus position, Children have been noted to experi- because air will have already ence greater hemodynamic derange- passed into the outflow tract by the ments from venous air embolism time of recognition. Turning the pathan adults. This is believed to be tient to right-side-up shifts the bodue to the volume of the air embolus of air into the right pulmonary lus being relatively large compared artery and may produce significant to their intracardiac volume (11). cardiovascular instability (4). Pathophysiologically, the effect of More critical than knowing how a large air embolism is the producto treat venous air embolism is taktion of a mechanical air block that ing measures to prevent it. The poobstructs right ventricular outflow. tential for venous air embolism is This leads to elevated right ventric- present whenever there is an open ular and pulmonary pressures conconduit into a vein under negative current with a decrease in left venpressure relative to atmospheric tricular and aortic pressures. pressure. Negative venous pressure Equally important is the blood-air is generated with negative intrathointeraction promoting platelet agracic pressure produced by spontagregation and fibrin formation, furneous respiration, creating the necther restricting blood flow. There is essary pressure gradient to entrain a release of vasoactive substances air into the central venous system. including histamine and serotonin, Adequate hydration to optimize resulting in pulmonary vasoconcentral venous pressure and use of striction, increased capillary perme- the Trendelenburg position attempt ability with pulmonary edema, and to decrease the pressure gradient increased airway resistance (12,131. between the open sheath and the Additionally, the increased right central venous system and, thereby, atrial pressure may cause arterial reduce the possibility of air emboembolization through a patent fora- lus. If the angiographic table cannot be placed in the Trendelenburg pomen ovale (14). Clinical signs and symptoms of sition, elevating the patient's legs venous air embolism are not specific will help increase central venous for this entity and may mimic other pressure. Also, pinching the open acute cardiorespiratory or cerebrosheath and attempting to exchange vascular events. Symptoms may in- the dilator for a catheter between clude altered mental status or loss breaths reduces the time the sheath of consciousness, seizure, neurologic is open to atmospheric pressure. deficit, and nausea and vomiting. These measures were followed but Typical signs of venous air embofailed to prevent venous air embolism are hypotension, tachypnea, lism in our three patients. and cyanosis. On cardiac auscultaThe most important measure is tion, a classic "mill-wheel" murmur to ensure that the patient does not may be heard along the left sternal take a spontaneous breath a t the border. This is dependent on a large critical time when the central veamount of air being present in the nous system is open to atmospheric right ventricle (10). pressure. In consideration of this, The treatment goal in venous air emphasis should be placed on preembolism is to restore flow within vention of negative intrathoracic the cardiovascular circulation. Mea- pressure. Adults or older children sures include patient positioning, who are not sedated, or are under removal of air from the venous cir"conscious sedation," are able to folculation, and CPR. Placing the palow commands to breath hold or tient with the left side down and in perform a Valsalva maneuver and, Trendelenburg (the Durant) posithus, maintain a positive intrathou
784
Air Embolism during Tunneled Central Catheter Placement
June 1999 JVIR
racic pressure. In smaller children this is not possible. It should be emphasized that in pediatrics, seldom what we do could be described as "conscious sedation." In virtually all procedures of this type, young children require "deep sedation," and, therefore, the patient is unable to follow commands such as breath holding. GA with PPV should be used in those children who cannot be expected to follow commands to suspend respirations. The use of GA ensures the patient will not create negative intrathoracic pressure by breathing spontaneously. It should be remembered that GA without PPV, however, is no different than deep sedation if the patient is not phannacologically paralyzed and is allowed to breathe spontaneously. Continuous PPV or aDnea is essential. Our experience with vascular access has grown steadily during the past 5 years, beginning with a series of 222 peripherally inserted central catheters (16). Our service has ex~andedduring this time to include tunneled ceitral venous catheters, temporary and permanent hemodialysis catheters, and subcutaneous ports. These three cases occurred in the first 100 procedures of tunneled central catheter placement. Recognition of the possible complication of venous air embolism during the critical step and the encounter of one near disaster (case 3) has prompted us to use GA with PPV in all small children (generally younger than 5 years of age).
The placement of a tunneled central catheter requires a critical step in which the central venous system is open to atmospheric pressure. It is imperative that negative intrathoracic pressure be avoided during this step. Children under deep sedation are unable to follow commands to suspend respiration and are a t increased risk for venous air embolism. Consideration should be given for using GA with PPV during tunneled central catheter placement in these cases. References 1. Denny DF. The role of the radiologist in long-term central-vein access. Radiology 1992; 185:637-638. 2. Committee on Drugs. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures. Pediatrics 1992; 89:lllO-1115. 3. Cote CJ. Sedation for the pediatric patient: a review. Pediatr Clin North Am 1994; 41:31-51. 4. Mauro MA, Jaques PF. Radiologic placement of Hickman catheters. JVIR 1993; 4:127-137. 5. DeCandia M, Bodner LJ. Tunneled central venous access catheter placement in the pediatric population: comparison of radiologic and surgical results. Radiology 1994; 192:265-268. 6. Lee JD, Roy TM. Venous air embolism and the central venous catheter. J Tenn Med Assoc 1995; 88:5153. 7. Lambert MJ. Air embolism in central venous catheterization: diagno-
sis, treatment, and prevention. South Med J 1982; 75:1189-1191. 8. Flanagan PJ, Gradisar IA, Gross RJ, et al. Air embolus: a lethal complication of subclavian venipuncture. N Engl J Med 1967; 281:488489. 9. Palmon SC, Moore LE, Lundberg J , et al. Venous air embolism: a review. J Clin Anesth 1997; 9:251257. 10. Orebaugh SL. Venous air embolism: clinical and experimental considerations. Crit Care Med 1992; 20:1169-1177. 11. Leicht CH, Waldman J. Pulmonary air embolism in the pediatric patient undergoing central catheter placement: a report of two cases. Anesthesiology 1986; 64:521-523. 12. O'Quin RJ, Lakshminarayan S. Venous air embolism. Arch Intern Med 1982; 142:2173-2176. 13. Adornato DC, Gildenberg PL, Ferrario CM, et al. Pathophysiology of intravenous air embolism in dogs. Anesthesiology 1978; 49:120-127. 14. Gottdiener JS, Papademetriou V, Notargiacomo A, et al. Incidence and cardiac effects of systemic venous air embolization via noncardiac shunt. Arch Intern Med 1988; 148:795-800. 15. Durant TM, Oppenheimer MJ, Lynch PR, et al. Body position in relation to venous air embolism: a roentgenologic study. Am J Med Sci 1954; 227:509-520. 16. Donaldson JS, Morello FP, Junewick J J , O'Donovan JC, LimDunham J. Peripherally inserted central venous catheters: US-guided vascular access in pediatric patients. Radiology 1995; 197:542-544.
Air Embolism during Tunneled Central Catheter Placement Performed without General Anesthesia in Children: A Potentially Serious Complication1 Frank P. Morello, MD James 5. Donaldson, MD Martha C. Saker, MD Jackson T. Norman, MD
Index terms: Catheters and catheterization, complications * Central venous access Embolism, air
JVIR 1999; 10:781-784 Abbreviations: CPR = cardiopulmonary resuscitation, GA = general anesthesia, IV = intravenous, PPV = positive pressure ventilation
From the Children's Memorial Medical Center, Northwestern University Medical School, 2300 Children's Plaza, Chicago, IL 60614. Received September 9, 1998; revision requested October 19; revision received December 21; accepted December 28. Address correspondence to F.P.M. 0 SCVIR, 1999
CENTRAL venous catheters have had an increasingly important role in a variety of patient care situations, including long-term antibiotic therapy, chemotherapy, and nutritional support. The recent past has seen a gradual transition from placement of vascular access catheters by surgeons to placement by interventional radiologists (1).The interventional radiology service places a majority of the vascular access devices at our children's hospital, including peripherally inserted central catheters, tunneled central venous catheters, temporary and permanent hemodialysis catheters, and subcutaneous ports. Most procedures performed by our interventional radiology service in children can be successfully completed with use of intravenous (IV) sedation, and a few require general anesthesia (GA). Key advantages of GA over IV sedation include the ability to have positive pressure ventilation (PPV) or controlled apnea during the procedure. We report our experience of venous air embolism in three small children during placement of tunneled central venous catheters when GA was not used.
1 TECHNIQUE The guidelines written by the Committee on Drugs of the American Academy of Pediatrics were followed for all procedures requiring sedation (2,3). While sedated, all
patients had continuous pulse oximetry monitoring and vital signs including blood pressure, heart rate, and respiration rate were monitored and recorded every 5 minutes. The technique used for tunneled central catheter la cement in children was identical t o techniques described for adults (4.5). . , All venipunctures were done with use of ultrasound guidance and micropuncture access (Cook, Bloomington, IN) into the right internal jugular vein. As a modification to the standard technique, a subcutaneous tunnel was then formed between the catheter exit and the venipuncture sites using a blunt tip needle (Hawkins needle; Cook). A guide wire was passed through the needle and, after the needle was removed, a peel-away sheath was advanced through the tunnel to facilitate passage of the catheter. This peel-away sheath was then removed, leaving the catheter in the subcutaneous tunnel. The inner dilator and wire of the micropuncture set in the internal jugular vein were exchanged for a 0.035-inch guide wire, and the outer dilator was then exchanged for an appropriately sized peel-away sheath, which was inserted into the vein (Fig 1).The next step was the critical one, where the inner dilator and wire were removed from the peel-away sheath temporarily exposing the open sheath to open air (Fig 2). The peel-away sheath was immediately pinched after removing ,
782
Air Embolism during Tunneled Central Catheter Placement
June 1999 JVIR
Figures 1, 2. (1) Placement of a tunneled central venous catheter. After venipuncture of the right internal jugular vein, a
peel-away sheath with inner dilator is advanced over a guide wire into the superior vena cava. The central venous system is not open to air. The catheter has been advanced through the subcutaneous tunnel. (2) The inner dilator and wire have been removed so that the catheter may be advanced into the peel-away sheath. The central venous system is open to air. the inner dilator and wire, and the catheter was quickly inserted through the open sheath into the superior vena cava, with the sheath being pinched between each forward push of the catheter. After catheter insertion, the peel-away sheath was removed, the catheter was sutured, and a sterile dressing was applied.
I CASE REPORTS Case 1.-A 3-year-old boy with cerebral palsy and developmental delay had osteomyelitis of the left tibia. Placement of a central catheter was needed for long-term IV antibiotic administration. The patient was not sedated for the procedure because of thick copious orotracheal secretions and because the patient did not have purposeful movements. The right internal jugular vein was used for access and our described technique for tunneled central catheter placement was followed. During the critical step, which is the very short time when the 5-F sheath was open to atmospheric pressure, the patient began to cough. This was thought to be due to his secretions. A soft hissing sound through the sheath was heard and the patient became hy-
poxic (02saturation, 85%) and bradycardic (heart rate, 60). Supplemental oxygen was administered and the child promptly recovered to baseline vital signs. The 4.2-F (Bard, Salt Lake City, UT) catheter was successfully placed as the patient recovered. Case 2.-An 18-month-old boy with end-stage renal disease secondary to posterior urethral valves needed a tunneled hemodialysis catheter while awaiting a renal transplant. The patient was sedated with titrated doses of ketamine and midazolam but remained slightly agitated throughout the procedure. The patient had undergone previous interventional radiologic procedures and was quite difficult to sedate. He would remain agitated at dose levels that would be adequate for other children, but be hypotensive a t higher levels. After placement of a 10-F peel-away sheath into the internal jugular vein, a 9-F catheter (Medcomp, Harleysville, PA) was being advanced into the sheath when the patient began crying. A hissing sound was heard around the catheter, which was already in the sheath. This corresponded to a 10% drop in oxygen saturation. The patient recovered quickly and reu
mained stable with supplemental oxygen as the procedure was completed. Case 3.-A 10-month-old girl with congenital leukemia was referred for central catheter placement for chemotherapy and future bone marrow transplant. She was sedated with titrated doses of ketamine and midazolam. During the critical step, as the 7-F double-lumen catheter (Bard, Salt Lake City, UT) was about to be advanced through the open 8-F sheath, the patient became irritable and took a deep spontaneous breath. Air was heard being loudly sucked into the sheath. The patient immediately became hypotensive and profoundly hypoxic with ensuing cardiorespiratory arrest. After vigorous cardiopulmonary resuscitation (CPR), including intubation, the patient was stabilized. During initial CPR and intubation, the sheath was held tightly pinched by the radiologist. After the patient was intubated, chest compressions were briefly interrupted while the central catheter was advanced into the sheath, and the sheath was then removed. Resuscitative efforts continued, with the central catheter being used to administer drugs and fluids. The
Morello et a1
783
Volume 10 Number 6
patient had no sequelae from this episode.
I DISCUSSION In the three cases reported, venous air embolism was a complication of tunneled central venous catheter placement. All three incidents occurred under similar circumstances. That is, all patients were breathing spontaneously during a critical step of the catheter insertion. A centrally placed sheath was temporarily exposed to open air during exchange of a dilator for the catheter. None of the patients could hold their breath on command (two being deeply sedated, and one a developmentally delayed child who was not sedated). Spontaneous breaths generated sufficient negative pressure to cause venous air embolism during the critical catheter exchange. Venous air embolism in two of the children were minor, but a near disaster occurred in the third child. The effects of venous air embolism can vary from clinically insignificant to com~letecardiovascular collapse. Venous air embolism can be the most dramatic and, potentially, the most serious complication of central venous catheters, with a reported mortality of 29%-36% (6). The volume of air tolerated is unknown but accidental injections of air of between 100 mL and 300 mL have been reported to be fatal (7). This volume becomes relevant considering that it has been calculated that 100 mL of air per second can flow through a 14-gauge needle, with only a 5 cm H,O pressure gradient (8).A 14-gauge needle has an inner diameter slightly larger than a 6-F sheath. The sheath sizes used in our case reports were 5 F, 8 F, and 10 F. A 10-F sheath has a cross-sectional area 2.7 times larger than that of a 6-F sheath. One can easily see how a large air bolus can be entrained through these vascular sheaths. The factors that determine the morbidity and mortality of venous air embolism include the volume of
air and the rate at which air enters tion allows air in the pulmonary the venous system, as well as the outflow tract to migrate to the apex position of the patient at the time of the right ventricle and permits the venous air embolism occurred flow of blood to the lungs (15). Oth(9,lO). Intrinsically, the adaptability ers have recommended leaving the of the patient's cardiovascular syspatient supine, rather than turning tem is probably a factor as well. to a left lateral decubitus position, Children have been noted to experi- because air will have already ence greater hemodynamic derange- passed into the outflow tract by the ments from venous air embolism time of recognition. Turning the pathan adults. This is believed to be tient to right-side-up shifts the bodue to the volume of the air embolus of air into the right pulmonary lus being relatively large compared artery and may produce significant to their intracardiac volume (11). cardiovascular instability (4). Pathophysiologically, the effect of More critical than knowing how a large air embolism is the producto treat venous air embolism is taktion of a mechanical air block that ing measures to prevent it. The poobstructs right ventricular outflow. tential for venous air embolism is This leads to elevated right ventric- present whenever there is an open ular and pulmonary pressures conconduit into a vein under negative current with a decrease in left venpressure relative to atmospheric tricular and aortic pressures. pressure. Negative venous pressure Equally important is the blood-air is generated with negative intrathointeraction promoting platelet agracic pressure produced by spontagregation and fibrin formation, furneous respiration, creating the necther restricting blood flow. There is essary pressure gradient to entrain a release of vasoactive substances air into the central venous system. including histamine and serotonin, Adequate hydration to optimize resulting in pulmonary vasoconcentral venous pressure and use of striction, increased capillary perme- the Trendelenburg position attempt ability with pulmonary edema, and to decrease the pressure gradient increased airway resistance (12,131. between the open sheath and the Additionally, the increased right central venous system and, thereby, atrial pressure may cause arterial reduce the possibility of air emboembolization through a patent fora- lus. If the angiographic table cannot be placed in the Trendelenburg pomen ovale (14). Clinical signs and symptoms of sition, elevating the patient's legs venous air embolism are not specific will help increase central venous for this entity and may mimic other pressure. Also, pinching the open acute cardiorespiratory or cerebrosheath and attempting to exchange vascular events. Symptoms may in- the dilator for a catheter between clude altered mental status or loss breaths reduces the time the sheath of consciousness, seizure, neurologic is open to atmospheric pressure. deficit, and nausea and vomiting. These measures were followed but Typical signs of venous air embofailed to prevent venous air embolism are hypotension, tachypnea, lism in our three patients. and cyanosis. On cardiac auscultaThe most important measure is tion, a classic "mill-wheel" murmur to ensure that the patient does not may be heard along the left sternal take a spontaneous breath a t the border. This is dependent on a large critical time when the central veamount of air being present in the nous system is open to atmospheric right ventricle (10). pressure. In consideration of this, The treatment goal in venous air emphasis should be placed on preembolism is to restore flow within vention of negative intrathoracic the cardiovascular circulation. Mea- pressure. Adults or older children sures include patient positioning, who are not sedated, or are under removal of air from the venous cir"conscious sedation," are able to folculation, and CPR. Placing the palow commands to breath hold or tient with the left side down and in perform a Valsalva maneuver and, Trendelenburg (the Durant) posithus, maintain a positive intrathou
784
Air Embolism during Tunneled Central Catheter Placement
June 1999 JVIR
racic pressure. In smaller children this is not possible. It should be emphasized that in pediatrics, seldom what we do could be described as "conscious sedation." In virtually all procedures of this type, young children require "deep sedation," and, therefore, the patient is unable to follow commands such as breath holding. GA with PPV should be used in those children who cannot be expected to follow commands to suspend respirations. The use of GA ensures the patient will not create negative intrathoracic pressure by breathing spontaneously. It should be remembered that GA without PPV, however, is no different than deep sedation if the patient is not phannacologically paralyzed and is allowed to breathe spontaneously. Continuous PPV or aDnea is essential. Our experience with vascular access has grown steadily during the past 5 years, beginning with a series of 222 peripherally inserted central catheters (16). Our service has ex~andedduring this time to include tunneled ceitral venous catheters, temporary and permanent hemodialysis catheters, and subcutaneous ports. These three cases occurred in the first 100 procedures of tunneled central catheter placement. Recognition of the possible complication of venous air embolism during the critical step and the encounter of one near disaster (case 3) has prompted us to use GA with PPV in all small children (generally younger than 5 years of age).
The placement of a tunneled central catheter requires a critical step in which the central venous system is open to atmospheric pressure. It is imperative that negative intrathoracic pressure be avoided during this step. Children under deep sedation are unable to follow commands to suspend respiration and are a t increased risk for venous air embolism. Consideration should be given for using GA with PPV during tunneled central catheter placement in these cases. References 1. Denny DF. The role of the radiologist in long-term central-vein access. Radiology 1992; 185:637-638. 2. Committee on Drugs. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures. Pediatrics 1992; 89:lllO-1115. 3. Cote CJ. Sedation for the pediatric patient: a review. Pediatr Clin North Am 1994; 41:31-51. 4. Mauro MA, Jaques PF. Radiologic placement of Hickman catheters. JVIR 1993; 4:127-137. 5. DeCandia M, Bodner LJ. Tunneled central venous access catheter placement in the pediatric population: comparison of radiologic and surgical results. Radiology 1994; 192:265-268. 6. Lee JD, Roy TM. Venous air embolism and the central venous catheter. J Tenn Med Assoc 1995; 88:5153. 7. Lambert MJ. Air embolism in central venous catheterization: diagno-
sis, treatment, and prevention. South Med J 1982; 75:1189-1191. 8. Flanagan PJ, Gradisar IA, Gross RJ, et al. Air embolus: a lethal complication of subclavian venipuncture. N Engl J Med 1967; 281:488489. 9. Palmon SC, Moore LE, Lundberg J , et al. Venous air embolism: a review. J Clin Anesth 1997; 9:251257. 10. Orebaugh SL. Venous air embolism: clinical and experimental considerations. Crit Care Med 1992; 20:1169-1177. 11. Leicht CH, Waldman J. Pulmonary air embolism in the pediatric patient undergoing central catheter placement: a report of two cases. Anesthesiology 1986; 64:521-523. 12. O'Quin RJ, Lakshminarayan S. Venous air embolism. Arch Intern Med 1982; 142:2173-2176. 13. Adornato DC, Gildenberg PL, Ferrario CM, et al. Pathophysiology of intravenous air embolism in dogs. Anesthesiology 1978; 49:120-127. 14. Gottdiener JS, Papademetriou V, Notargiacomo A, et al. Incidence and cardiac effects of systemic venous air embolization via noncardiac shunt. Arch Intern Med 1988; 148:795-800. 15. Durant TM, Oppenheimer MJ, Lynch PR, et al. Body position in relation to venous air embolism: a roentgenologic study. Am J Med Sci 1954; 227:509-520. 16. Donaldson JS, Morello FP, Junewick J J , O'Donovan JC, LimDunham J. Peripherally inserted central venous catheters: US-guided vascular access in pediatric patients. Radiology 1995; 197:542-544.