ttp://www.bsava.com
PAPER
Venous air embolism detected on computed tomography of small animals H. G. Heng*, J. D. Ruth*† and K. Lee‡ *Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Purdue University, West Lafayette, IN 47906, USA †Department of Small Animal Clinical Sciences, Virginia-Maryland Regional College of Veterinary Medicine, Blacksburg, VA 24061, USA ‡College of Veterinary Medicine, Chonbuk National University, Jeonju 561-756, Republic of Korea
OBJECTIVE: To describe the prevalence, location and clinical significance of abnormal gas accumulations in dogs and cats detected on computerised tomography images. METHODS: Retrospective evaluation of all canine and feline computed tomography examinations (292 pre-contrast and 219 post-contrast) performed in a 12-month time period. All studies were evaluated for the presence of venous air emboli. The location of intravenous gas was noted and the volume of intravenous air emboli was estimated visually. The medical records of animals with venous air embolism were reviewed for signs of cardiopulmonary complications. RESULTS: The overall prevalence of air embolism on pre- and incidence on post-contrast images was 4·5 and 2·3%, respectively. The prevalence of air embolism on pre-contrast and incidence on post-contrast thoracic images was 35·7 and 14·2%, respectively. The volume of venous air was generally small and the most common was in an axillary vein. None of the animals had any cardiopulmonary complications. CLINICAL SIGNIFICANCE: The presence of small volume venous air embolism on routine computed tomography examinations is a frequent incidental finding that does not appear to cause cardiopulmonary complications.
Journal of Small Animal Practice (2014) 55, 420–423 DOI: 10.1111/jsap.12238 Accepted: 22 April 2014; Published online: 2 June 2014
INTRODUCTION Introduction of venous air emboli is a common occurrence during vessel catheterisation (Groell et al. 1997a, Boitout & Mahler 2013) and contrast administration for human computed tomography (CT) examinations (Price et al. 1987, Groell et al. 1997b, Sakai et al. 1998, Sodhi et al. 2012). In addition, various medical and surgical procedures have been reported to lead to air embolism (Ackerman et al. 1972, Thayer et al. 1980, Nern et al. 2012, Riddick & Brogdon 2012, Pandey et al. 2013). In humans, the presence of a small venous air embolism is usually considered an incidental finding, without resultant complications (Woodring & Fried 1988, Rubinstein et al. 1996). If a large amount of venous gas is present, cardiac arrest (Pacifico et al. 2010), respiratory arrest (Boitout & Mahler 2013) and eventually death (Ackerman et al. 1972, Thayer et al. 1980, Ober et al. 2006, 420
Riddick & Brogdon 2012, Moningi et al. 2013, Sopena-Falco et al. 2013) can occur. Venous air emboli have been reported in 11·7 to 23% of human patients undergoing contrast-enhanced CT examinations (Woodring & Fried 1988, Groell et al. 1997b). Similarly, the authors have noted venous air emboli in canine and feline patients during routine CT examinations (unpublished). The prevalence of pre-contrast CT and incidence of post-contrast CT of venous air embolism, and its clinical significance is not well documented in dogs and cats. The objective of this study was to determine the prevalence of pre-contrast, incidence of post-contrast, location and clinical significance of venous air embolism in dogs and cats when detected on CT. It was hypothesised that canine and feline venous air embolism is an incidental finding that occurs with similar frequency as reported in humans.
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
© 2014 British Small Animal Veterinary Association
CT detection of venous air embolism
MATERIALS AND METHODS Computed tomography examinations of all canine and feline patients performed at Purdue University Veterinary Teaching Hospital from November 2010 to October 2011 were included in the study. All studies were performed on the same multi-detector CT scanner (Light Speed QX/I, GE Medical System). Scan parameters (slice thickness, pitch and algorithm) were optimised for each patient. Anaesthetic or sedative protocols varied depending on clinician preference, and patient position varied depending on the location of the region of the interest. Each study was reviewed by two radiologists (HGH and KCL) and evaluated for the presence of venous air embolism. For each study, the regions of the examination were recorded. When present, the location of intravenous gas was noted and the amount was estimated visually using categorical variables (small, moderate and large amount). Gas volume was classified as a small amount when it was only detected in a single transverse image, moderate when the venous gas was detected in two to three consecutive transverse images and large when it could be detected in four or more consecutive images. When present, it was noted whether the intravenous gas was detected on the pre-contrast examination, post-contrast examination or both. The medical records of the included cases were reviewed to determine if any cardiopulmonary complication (cardiac arrest, respiratory distress or respiratory arrest) occurred during and/or after the CT examination.
RESULTS A total of 292 pre-contrast and 219 post-contrast CT studies were reviewed, including 256 dogs and 36 cats. The regions examined included the head (n=140; 118 dogs, 22 cats), spine (n=70; 66 dogs, 4 cats), abdomen (n=38; 34 dogs, 4 cats), thorax (n=28; 26 dogs, 2 cats) and extremity (n=16; 12 dogs, 4 cats). Venous air embolism was detected in a total of 18 cases (16 dogs and 2 cats), two of which were not administered intravenous contrast (3 in only pre-contrast, 5 in only post-contrast and 10 in both pre- and post-contrast). Fourteen of the 18 studies evaluated the thoracic region, three evaluated the cervical spine and one evaluated the abdomen (Table 1). Sixteen animals were positioned in sternal recumbency, and two in dorsal recumbency. All thoracic studies had pre- and post-contrast studies. No venous air embolism was detected on head or extremity studies. The overall prevalence of pre-contrast venous air embolism was 4·5% (13 of 292). The incidence of air embolism subsequent to intravenous injection Table 1. Region of the CT study where venous air embolism was detected during the pre- and post-contrast study Region of CT study
Precontrast
Postcontrast
Only precontrast
Only postcontrast
Thorax Spine Abdomen Total
10 2 1 13
11 3 1 15
3 0 0 3
4 1 0 5
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
Both pre- and post-contrast 7 2 1 10
FIG 1. Pre- (A) and post-contrast (B) images of a dog. Small amount of intravenous air embolus (white arrow) in the caudal vena cava was only present in the pre-contrast thoracic CT study
FIG 2. Pre- (A) and post-contrast (B) CT images of a dog. Intravenous air emboli were present in the right axillobrachial vein (thin black arrows) and right external jugular vein (thick black arrows) on both pre and post contrast CT examinations. Note that the intravenous air did not change position between pre- and post-contrast examinations
of contrast was 2·3% (5 of 219). The prevalence of thoracic pre-contrast [35·7% (10 of 28)] and incidence of post-contrast [14·2% (4 of 28)] venous air embolism was calculated separately. Venous air embolism was most commonly (15 studies) seen in the axillary vein and less commonly in the brachiocephalic vein (one study), jugular vein (one study), femoral vein (one study), interarcuate vein (one study) and caudal vena cava (one study) (Fig 1). Two dogs had two locations of venous gas embolism in a single CT study, including the axillary and jugular veins (Fig 2), and the axillary and inter-arcuate veins. When venous air embolism was detected in both pre- and post-contrast CT studies, the location of gas did not change between studies (Fig 2). In pre-contrast CT studies, the estimated volume of the venous air embolism was small in four animals; moderate in four animals and large in five animals. In post-contrast CT studies, the number of animals having small, moderate and large amount of venous air embolism were four, seven and four, respectively. In the 10 animals with pre- and post-contrast venous air embolism, there was increased estimated venous air embolism volume on the post-contrast study in 6 animals. The volume was unchanged in three animals and reduced in one animal. None of the animals demonstrated any signs of cardiopulmonary complications.
DISCUSSION The major finding of this study is that detection of venous air embolism can be expected in canine and feline CT examinations as an incidental finding. Computed tomography is widely applied in human and veterinary medicine for investigation of
© 2014 British Small Animal Veterinary Association
421
H. G. Heng et al.
a variety of diseases (Pollard & Puchalski 2011). The majority of cases require intravenous catheterisation for administration of intravenous fluids, injectable anaesthetic agents and contrast media (Pollard & Puchalski 2011). When performing contrast enhanced CT in humans, venous air embolism is encountered as a non-fatal incidental finding (Price et al. 1987, Woodring & Fried 1988, Rubinstein et al. 1996, Groell et al. 1997b, Sakai et al. 1998, Sodhi et al. 2012), despite attempts to reduce or prevent introducing venous gas. The majority of the CT examinations included in this study evaluated the head and spine (72%) and this region of interest excluded the most common site of venous embolism (axillary vein), thus the true prevalence of venous air embolism may be underestimated. As a result, the reported prevalence of venous air embolism seen on thoracic CT (35·7%) may accurately reflect the true prevalence. It is proposed that the venous air embolism observed before intravenous contrast administration was introduced during intravenous catheterisation, flushing the catheter with heparinised saline, or administration of injectable anaesthetic agents (Cook 2013). As there is no specific difference in catheterisation technique and positioning, dogs and cats are expected to have equal prevalence of intravenous gas emboli. The prevalence of pre-contrast venous air embolism of the thoracic CT has been estimated as 5·5% in one human study (Groell et al. 1997b), lower than in this report on pre-contrast thoracic CT. One of the important factors that may have led to the higher prevalence in this study is that fact that all of the patients were under general anaesthesia or heavy sedation, necessitating multiple intravenous injections via an indwelling catheter. In addition, this is a teaching hospital where senior veterinary students and veterinary technician students are given the opportunity to perform some procedures and as a result, multiple catheterisation attempts may have occurred. The incidence of post-contrast CT intravenous air embolism reportedly ranges from 11·7 to 23% in humans (Groell et al. 1997b, Sakai et al. 1998). An incidence of 4% was reported in a prospective human study (Sakai et al. 1998). The lower incidence in the prospective study suggests that careful handling of butterfly needle; extension tube and power injector may reduce the risk of introducing venous air embolism (Sakai et al. 1998). Surprisingly a lower incidence of venous air embolism in post-contrast thoracic CT of 14·2% (4 of 28) was found in the present study when compared with pre-contrast thoracic CT, which is similar to that reported in humans. The large difference of the prevalence of the pre-contrast and incidence of the post-contrast study may be due to hospital policy requiring more experienced personnel (only certified veterinary technicians) to perform intravenous contrast injection. In humans, most CT contrast was injected using a power injector with extension tubing. At the time of this study the institution did not use a power injector or extension tubing, and contrast media was manually injected directly into a preplaced indwelling catheter. Both ways of contract injections appear to result in similar prevalence of venous air embolism. No adverse effects were noted secondary to the presence of venous air emboli, likely due to the relatively small volume. Morbidity and mortality rates from venous air embolism are determined by the position of the patient and the volume of air 422
introduced. Human patients with right to left intra-cardiac shuts or pulmonary arteriovenous malformation may have complications of cerebral air embolism (Woodring & Fried, 1988, Sodhi et al. 2012). This may also be true for dogs and cats. Therefore, one should be cautious when routinely administering intravenous fluids and medications in dogs and cats with these disease conditions. While patient position varied, the volume of the venous air embolism present in this report was subjectively judged to be less than 1 mL in all cases (estimated with the formula of volume = π×radius2×length). The increased estimated volume of air in postcontrast studies was probably due to air being introduced during the injection of the contrast media. Since repositioning was not performed between pre- and post-contrast studies, redistribution of the gas is a less likely cause of the increased estimated volume of air in the post-contrast study. The lethal dose of intravenous air in the dog has been estimated at 40 to 50 mL in a small dog and 60 to 200 mL in a large dog (Green 1904), or approximately 7·5 mL/kg (Oppenheimer et al. 1953). Iatrogenic fatal air embolism has been reported in animals undergoing procedures such as pneumocystography, cryosurgery and laparoscopy (Ackerman et al. 1972, Harvey 1978, Gilroy & Anson 1987). Non-fatal cardiac arrest and respiratory arrest in two cats has been reported secondary to accidental iatrogenic venous air embolisation (Pacifico et al. 2010, Boitout & Mahler 2013). Both of these two cats, survived due to immediate emergency response and treatment. Other causes of venous air embolism such as infection (Gaschen et al. 2003), trauma (Sen et al. 2013) and necrosis (Gaschen et al. 2003) should be ruled out. Experience with intravenous catheterisation and injection might aid in preventing or reducing the chances of introducing intravenous air embolism. Similar to humans (forearm vein), the preferred site for catheterisation and intravenous injection in the authors’ institute is the cephalic vein of a thoracic limb. The location of intravenous air in this study is compatible with human studies where postcontrast CT intravenous gas was commonly seen in the subclavian or internal jugular veins (Rubinstein et al. 1996, Sakai et al. 1998). Detection of intravenous gas on post-contrast CT has also been described in the main pulmonary artery, cardiac chambers and superior vena cava (Woodring et al. 1988, Groell et al. 1997b, Sodhi et al. 2012), which was not seen in the animal studied here. This difference may be attributed to differences in catheterisation technique or volume of air introduced. In the two cases where air embolism was encountered caudal to the heart, (femoral vein and caudal vena cava) the catheterisation site was a saphenous vein. The presence of venous air embolism in the interarcuate vein is unusual and has not been reported. When blood flows from the cephalic and saphenous veins, air emboli will be trapped and absorbed in either the right heart chambers or pulmonary capillaries and should not be able to circulate in vascular system. Careful review of this case did not reveal a shunting vessel that could lead to embolisation of the interarcuate vein. No arterial air emboli were observed, likely because there were no cases of right to left shunt or arteriovenous malformation in the dogs and cats in this study. In conclusion, the clinical significance of venous air emboli detected by CT is questionable. Venous air emboli are often an
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
© 2014 British Small Animal Veterinary Association
CT detection of venous air embolism
incidental finding, without associated cardiopulmonary complications. However, the finding of intravenous injection related venous air emboli should be differentiated from other possible causes such as trauma, necrosis or infection. References Ackerman, N., Wingfield, W. E. & Corley, E. A. (1972) Fatal air embolism associated with pneumourethrography and pneumocystography in a dog. Journal of American Veterinary Medical Association 160, 1616-1618 Boitout, V. & Mahler, S. P. (2013) Non-fatal high-rate intragenic venous air embolism in a cat. Veterinary Anaesthesia and analgesia 40, 336-338 Cook, L. S. (2013) Infusion-related air embolism. Journal of Infusion Nursing 36, 26-36 Gilroy, B. A. & Anson, L. W. (1987) Fatal air embolism during anesthesia for laparoscopy in a dog. Journal of American Veterinary Medical Association 190, 552-554 Green, J. S. (1904) The presence of air in the veins as a cause of death. American Journal of Medical Sciences 128, 38-65 Groell, R., Schafler, G. & Rienmueller, R. (1997a) The peripheral intravenous cannula: A cause of venous air embolism. American Journal of Medical Sciences 314, 300-302 Groell, R., Schaffler, G. J., Reinmueller, R., et al. (1997b) Vascular air embolism: location, frequency, and cause on electron-beam CT studies of the chest. Radiology 202, 459-462 Harvey, J. H. (1978) Fatal air embolism associated with cryosurgery in two dogs. Journal of American Veterinary Medical Association 173, 175-176 Moningi, S., Sulkarni, D. & Bhattacharjee, S. (2013) Coagulopathy following venous air embolism: a disastrous consequence – a case report. Korean Journal of Anesthesiology 65, 349-352 Nern, C., Bellut, B., Husain, N., et al. (2012) Fatal cerebral venous air embolism during endoscopic retrograde cholangiopancreatography – case report and review of the literature. Clinical Neuroradiology 22, 371-374
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
Ober, C. P., Sportwood, T. C., Hancock, R. (2006) Fatal venous air embolism in a cat with a retropharyngeal diverticulum. Veterinary Radiology & Ultrasound 47, 153-158 Oppenheimer, M. J., Durant, T. M., Lynch, P. (1953) Body position relation to venous air embolism and the associated cardiovascular-respiratory changes. American Journal of Medical Sciences 225, 362-373 Pacifico, N., Weishaar, K. M., Boozer, L. B., et al. (2010) Full recovery after cardiac arrest secondary to accidental iatrogenic venous air embolism in a cat. Journal of Veterinary Emergency and Critical Care 20, 264-267 Pandey, V., Varghese, E., Rao, M., et al. (2013) Nonfatal air embolism during shoulder arthroscopy. American Journal of Orthopedics 42, 272-274 Pollard, R. & Puchalski, S. (2011) CT contrast media and applications. In: Veterinary Computed Tomography. Eds T. Schwarz and J. Saunders. WileyBlackwell, Ames, IA, USA. pp 57-65 Price, D. B., Nardi, P. & Teitcher, J. (1987) Venous air embolization as a complication of pressure injection of contrast media: CT findings. Journal of Computer Assisted Tomography 11, 294-295 Riddick, L. & Brogdon, B. G. (2012) Fatal air embolism during renal dialysis. American Journal of Forensic Medicine and Pathology 33, 110-112 Rubinstein, D., Dangleis, K., Damiano, T. R. (1996) Venous air emboli identified on head and neck CT scans. Journal of Computer Assisted Tomography 20, 559-562 Sakai, O., Nakashima, N., Shinozaki, T., et al. (1998) Air bubbles in the subclavian or internal jugular veins: a common finding on contrast-enhanced CT. Neuroradiology 40, 258-260 Sodhi, K. S., Das, P. J. & Malhotra, P. (2012) Venous air embolism after intravenous contrast administration for computed tomography. The Journal of Emergency Medicine 42, 450-451 Sopena-Falco, J., Poch-Vall, N., Brullet, E., et al. (2013) Fatal massive air embolism following diagnostic colonoscopy. Endoscopy 45(Suppl 2), E91 Thayer, G. W., Carrig, C. B., Evans, A. T. (1980) Fatal venous air embolism associated with pneumocystography in a cat. Journal of American Veterinary Medical Association 176, 643-645 Woodring, J. H. & Fried, A. M. (1988) Nonfatal venous air embolism after contrastenhanced CT. Radiology 167, 405-407
© 2014 British Small Animal Veterinary Association
423
PAPER
Venous air embolism detected on computed tomography of small animals H. G. Heng*, J. D. Ruth*† and K. Lee‡ *Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Purdue University, West Lafayette, IN 47906, USA †Department of Small Animal Clinical Sciences, Virginia-Maryland Regional College of Veterinary Medicine, Blacksburg, VA 24061, USA ‡College of Veterinary Medicine, Chonbuk National University, Jeonju 561-756, Republic of Korea
OBJECTIVE: To describe the prevalence, location and clinical significance of abnormal gas accumulations in dogs and cats detected on computerised tomography images. METHODS: Retrospective evaluation of all canine and feline computed tomography examinations (292 pre-contrast and 219 post-contrast) performed in a 12-month time period. All studies were evaluated for the presence of venous air emboli. The location of intravenous gas was noted and the volume of intravenous air emboli was estimated visually. The medical records of animals with venous air embolism were reviewed for signs of cardiopulmonary complications. RESULTS: The overall prevalence of air embolism on pre- and incidence on post-contrast images was 4·5 and 2·3%, respectively. The prevalence of air embolism on pre-contrast and incidence on post-contrast thoracic images was 35·7 and 14·2%, respectively. The volume of venous air was generally small and the most common was in an axillary vein. None of the animals had any cardiopulmonary complications. CLINICAL SIGNIFICANCE: The presence of small volume venous air embolism on routine computed tomography examinations is a frequent incidental finding that does not appear to cause cardiopulmonary complications.
Journal of Small Animal Practice (2014) 55, 420–423 DOI: 10.1111/jsap.12238 Accepted: 22 April 2014; Published online: 2 June 2014
INTRODUCTION Introduction of venous air emboli is a common occurrence during vessel catheterisation (Groell et al. 1997a, Boitout & Mahler 2013) and contrast administration for human computed tomography (CT) examinations (Price et al. 1987, Groell et al. 1997b, Sakai et al. 1998, Sodhi et al. 2012). In addition, various medical and surgical procedures have been reported to lead to air embolism (Ackerman et al. 1972, Thayer et al. 1980, Nern et al. 2012, Riddick & Brogdon 2012, Pandey et al. 2013). In humans, the presence of a small venous air embolism is usually considered an incidental finding, without resultant complications (Woodring & Fried 1988, Rubinstein et al. 1996). If a large amount of venous gas is present, cardiac arrest (Pacifico et al. 2010), respiratory arrest (Boitout & Mahler 2013) and eventually death (Ackerman et al. 1972, Thayer et al. 1980, Ober et al. 2006, 420
Riddick & Brogdon 2012, Moningi et al. 2013, Sopena-Falco et al. 2013) can occur. Venous air emboli have been reported in 11·7 to 23% of human patients undergoing contrast-enhanced CT examinations (Woodring & Fried 1988, Groell et al. 1997b). Similarly, the authors have noted venous air emboli in canine and feline patients during routine CT examinations (unpublished). The prevalence of pre-contrast CT and incidence of post-contrast CT of venous air embolism, and its clinical significance is not well documented in dogs and cats. The objective of this study was to determine the prevalence of pre-contrast, incidence of post-contrast, location and clinical significance of venous air embolism in dogs and cats when detected on CT. It was hypothesised that canine and feline venous air embolism is an incidental finding that occurs with similar frequency as reported in humans.
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
© 2014 British Small Animal Veterinary Association
CT detection of venous air embolism
MATERIALS AND METHODS Computed tomography examinations of all canine and feline patients performed at Purdue University Veterinary Teaching Hospital from November 2010 to October 2011 were included in the study. All studies were performed on the same multi-detector CT scanner (Light Speed QX/I, GE Medical System). Scan parameters (slice thickness, pitch and algorithm) were optimised for each patient. Anaesthetic or sedative protocols varied depending on clinician preference, and patient position varied depending on the location of the region of the interest. Each study was reviewed by two radiologists (HGH and KCL) and evaluated for the presence of venous air embolism. For each study, the regions of the examination were recorded. When present, the location of intravenous gas was noted and the amount was estimated visually using categorical variables (small, moderate and large amount). Gas volume was classified as a small amount when it was only detected in a single transverse image, moderate when the venous gas was detected in two to three consecutive transverse images and large when it could be detected in four or more consecutive images. When present, it was noted whether the intravenous gas was detected on the pre-contrast examination, post-contrast examination or both. The medical records of the included cases were reviewed to determine if any cardiopulmonary complication (cardiac arrest, respiratory distress or respiratory arrest) occurred during and/or after the CT examination.
RESULTS A total of 292 pre-contrast and 219 post-contrast CT studies were reviewed, including 256 dogs and 36 cats. The regions examined included the head (n=140; 118 dogs, 22 cats), spine (n=70; 66 dogs, 4 cats), abdomen (n=38; 34 dogs, 4 cats), thorax (n=28; 26 dogs, 2 cats) and extremity (n=16; 12 dogs, 4 cats). Venous air embolism was detected in a total of 18 cases (16 dogs and 2 cats), two of which were not administered intravenous contrast (3 in only pre-contrast, 5 in only post-contrast and 10 in both pre- and post-contrast). Fourteen of the 18 studies evaluated the thoracic region, three evaluated the cervical spine and one evaluated the abdomen (Table 1). Sixteen animals were positioned in sternal recumbency, and two in dorsal recumbency. All thoracic studies had pre- and post-contrast studies. No venous air embolism was detected on head or extremity studies. The overall prevalence of pre-contrast venous air embolism was 4·5% (13 of 292). The incidence of air embolism subsequent to intravenous injection Table 1. Region of the CT study where venous air embolism was detected during the pre- and post-contrast study Region of CT study
Precontrast
Postcontrast
Only precontrast
Only postcontrast
Thorax Spine Abdomen Total
10 2 1 13
11 3 1 15
3 0 0 3
4 1 0 5
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
Both pre- and post-contrast 7 2 1 10
FIG 1. Pre- (A) and post-contrast (B) images of a dog. Small amount of intravenous air embolus (white arrow) in the caudal vena cava was only present in the pre-contrast thoracic CT study
FIG 2. Pre- (A) and post-contrast (B) CT images of a dog. Intravenous air emboli were present in the right axillobrachial vein (thin black arrows) and right external jugular vein (thick black arrows) on both pre and post contrast CT examinations. Note that the intravenous air did not change position between pre- and post-contrast examinations
of contrast was 2·3% (5 of 219). The prevalence of thoracic pre-contrast [35·7% (10 of 28)] and incidence of post-contrast [14·2% (4 of 28)] venous air embolism was calculated separately. Venous air embolism was most commonly (15 studies) seen in the axillary vein and less commonly in the brachiocephalic vein (one study), jugular vein (one study), femoral vein (one study), interarcuate vein (one study) and caudal vena cava (one study) (Fig 1). Two dogs had two locations of venous gas embolism in a single CT study, including the axillary and jugular veins (Fig 2), and the axillary and inter-arcuate veins. When venous air embolism was detected in both pre- and post-contrast CT studies, the location of gas did not change between studies (Fig 2). In pre-contrast CT studies, the estimated volume of the venous air embolism was small in four animals; moderate in four animals and large in five animals. In post-contrast CT studies, the number of animals having small, moderate and large amount of venous air embolism were four, seven and four, respectively. In the 10 animals with pre- and post-contrast venous air embolism, there was increased estimated venous air embolism volume on the post-contrast study in 6 animals. The volume was unchanged in three animals and reduced in one animal. None of the animals demonstrated any signs of cardiopulmonary complications.
DISCUSSION The major finding of this study is that detection of venous air embolism can be expected in canine and feline CT examinations as an incidental finding. Computed tomography is widely applied in human and veterinary medicine for investigation of
© 2014 British Small Animal Veterinary Association
421
H. G. Heng et al.
a variety of diseases (Pollard & Puchalski 2011). The majority of cases require intravenous catheterisation for administration of intravenous fluids, injectable anaesthetic agents and contrast media (Pollard & Puchalski 2011). When performing contrast enhanced CT in humans, venous air embolism is encountered as a non-fatal incidental finding (Price et al. 1987, Woodring & Fried 1988, Rubinstein et al. 1996, Groell et al. 1997b, Sakai et al. 1998, Sodhi et al. 2012), despite attempts to reduce or prevent introducing venous gas. The majority of the CT examinations included in this study evaluated the head and spine (72%) and this region of interest excluded the most common site of venous embolism (axillary vein), thus the true prevalence of venous air embolism may be underestimated. As a result, the reported prevalence of venous air embolism seen on thoracic CT (35·7%) may accurately reflect the true prevalence. It is proposed that the venous air embolism observed before intravenous contrast administration was introduced during intravenous catheterisation, flushing the catheter with heparinised saline, or administration of injectable anaesthetic agents (Cook 2013). As there is no specific difference in catheterisation technique and positioning, dogs and cats are expected to have equal prevalence of intravenous gas emboli. The prevalence of pre-contrast venous air embolism of the thoracic CT has been estimated as 5·5% in one human study (Groell et al. 1997b), lower than in this report on pre-contrast thoracic CT. One of the important factors that may have led to the higher prevalence in this study is that fact that all of the patients were under general anaesthesia or heavy sedation, necessitating multiple intravenous injections via an indwelling catheter. In addition, this is a teaching hospital where senior veterinary students and veterinary technician students are given the opportunity to perform some procedures and as a result, multiple catheterisation attempts may have occurred. The incidence of post-contrast CT intravenous air embolism reportedly ranges from 11·7 to 23% in humans (Groell et al. 1997b, Sakai et al. 1998). An incidence of 4% was reported in a prospective human study (Sakai et al. 1998). The lower incidence in the prospective study suggests that careful handling of butterfly needle; extension tube and power injector may reduce the risk of introducing venous air embolism (Sakai et al. 1998). Surprisingly a lower incidence of venous air embolism in post-contrast thoracic CT of 14·2% (4 of 28) was found in the present study when compared with pre-contrast thoracic CT, which is similar to that reported in humans. The large difference of the prevalence of the pre-contrast and incidence of the post-contrast study may be due to hospital policy requiring more experienced personnel (only certified veterinary technicians) to perform intravenous contrast injection. In humans, most CT contrast was injected using a power injector with extension tubing. At the time of this study the institution did not use a power injector or extension tubing, and contrast media was manually injected directly into a preplaced indwelling catheter. Both ways of contract injections appear to result in similar prevalence of venous air embolism. No adverse effects were noted secondary to the presence of venous air emboli, likely due to the relatively small volume. Morbidity and mortality rates from venous air embolism are determined by the position of the patient and the volume of air 422
introduced. Human patients with right to left intra-cardiac shuts or pulmonary arteriovenous malformation may have complications of cerebral air embolism (Woodring & Fried, 1988, Sodhi et al. 2012). This may also be true for dogs and cats. Therefore, one should be cautious when routinely administering intravenous fluids and medications in dogs and cats with these disease conditions. While patient position varied, the volume of the venous air embolism present in this report was subjectively judged to be less than 1 mL in all cases (estimated with the formula of volume = π×radius2×length). The increased estimated volume of air in postcontrast studies was probably due to air being introduced during the injection of the contrast media. Since repositioning was not performed between pre- and post-contrast studies, redistribution of the gas is a less likely cause of the increased estimated volume of air in the post-contrast study. The lethal dose of intravenous air in the dog has been estimated at 40 to 50 mL in a small dog and 60 to 200 mL in a large dog (Green 1904), or approximately 7·5 mL/kg (Oppenheimer et al. 1953). Iatrogenic fatal air embolism has been reported in animals undergoing procedures such as pneumocystography, cryosurgery and laparoscopy (Ackerman et al. 1972, Harvey 1978, Gilroy & Anson 1987). Non-fatal cardiac arrest and respiratory arrest in two cats has been reported secondary to accidental iatrogenic venous air embolisation (Pacifico et al. 2010, Boitout & Mahler 2013). Both of these two cats, survived due to immediate emergency response and treatment. Other causes of venous air embolism such as infection (Gaschen et al. 2003), trauma (Sen et al. 2013) and necrosis (Gaschen et al. 2003) should be ruled out. Experience with intravenous catheterisation and injection might aid in preventing or reducing the chances of introducing intravenous air embolism. Similar to humans (forearm vein), the preferred site for catheterisation and intravenous injection in the authors’ institute is the cephalic vein of a thoracic limb. The location of intravenous air in this study is compatible with human studies where postcontrast CT intravenous gas was commonly seen in the subclavian or internal jugular veins (Rubinstein et al. 1996, Sakai et al. 1998). Detection of intravenous gas on post-contrast CT has also been described in the main pulmonary artery, cardiac chambers and superior vena cava (Woodring et al. 1988, Groell et al. 1997b, Sodhi et al. 2012), which was not seen in the animal studied here. This difference may be attributed to differences in catheterisation technique or volume of air introduced. In the two cases where air embolism was encountered caudal to the heart, (femoral vein and caudal vena cava) the catheterisation site was a saphenous vein. The presence of venous air embolism in the interarcuate vein is unusual and has not been reported. When blood flows from the cephalic and saphenous veins, air emboli will be trapped and absorbed in either the right heart chambers or pulmonary capillaries and should not be able to circulate in vascular system. Careful review of this case did not reveal a shunting vessel that could lead to embolisation of the interarcuate vein. No arterial air emboli were observed, likely because there were no cases of right to left shunt or arteriovenous malformation in the dogs and cats in this study. In conclusion, the clinical significance of venous air emboli detected by CT is questionable. Venous air emboli are often an
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
© 2014 British Small Animal Veterinary Association
CT detection of venous air embolism
incidental finding, without associated cardiopulmonary complications. However, the finding of intravenous injection related venous air emboli should be differentiated from other possible causes such as trauma, necrosis or infection. References Ackerman, N., Wingfield, W. E. & Corley, E. A. (1972) Fatal air embolism associated with pneumourethrography and pneumocystography in a dog. Journal of American Veterinary Medical Association 160, 1616-1618 Boitout, V. & Mahler, S. P. (2013) Non-fatal high-rate intragenic venous air embolism in a cat. Veterinary Anaesthesia and analgesia 40, 336-338 Cook, L. S. (2013) Infusion-related air embolism. Journal of Infusion Nursing 36, 26-36 Gilroy, B. A. & Anson, L. W. (1987) Fatal air embolism during anesthesia for laparoscopy in a dog. Journal of American Veterinary Medical Association 190, 552-554 Green, J. S. (1904) The presence of air in the veins as a cause of death. American Journal of Medical Sciences 128, 38-65 Groell, R., Schafler, G. & Rienmueller, R. (1997a) The peripheral intravenous cannula: A cause of venous air embolism. American Journal of Medical Sciences 314, 300-302 Groell, R., Schaffler, G. J., Reinmueller, R., et al. (1997b) Vascular air embolism: location, frequency, and cause on electron-beam CT studies of the chest. Radiology 202, 459-462 Harvey, J. H. (1978) Fatal air embolism associated with cryosurgery in two dogs. Journal of American Veterinary Medical Association 173, 175-176 Moningi, S., Sulkarni, D. & Bhattacharjee, S. (2013) Coagulopathy following venous air embolism: a disastrous consequence – a case report. Korean Journal of Anesthesiology 65, 349-352 Nern, C., Bellut, B., Husain, N., et al. (2012) Fatal cerebral venous air embolism during endoscopic retrograde cholangiopancreatography – case report and review of the literature. Clinical Neuroradiology 22, 371-374
Journal of Small Animal Practice
•
Vol 55
•
August 2014
•
Ober, C. P., Sportwood, T. C., Hancock, R. (2006) Fatal venous air embolism in a cat with a retropharyngeal diverticulum. Veterinary Radiology & Ultrasound 47, 153-158 Oppenheimer, M. J., Durant, T. M., Lynch, P. (1953) Body position relation to venous air embolism and the associated cardiovascular-respiratory changes. American Journal of Medical Sciences 225, 362-373 Pacifico, N., Weishaar, K. M., Boozer, L. B., et al. (2010) Full recovery after cardiac arrest secondary to accidental iatrogenic venous air embolism in a cat. Journal of Veterinary Emergency and Critical Care 20, 264-267 Pandey, V., Varghese, E., Rao, M., et al. (2013) Nonfatal air embolism during shoulder arthroscopy. American Journal of Orthopedics 42, 272-274 Pollard, R. & Puchalski, S. (2011) CT contrast media and applications. In: Veterinary Computed Tomography. Eds T. Schwarz and J. Saunders. WileyBlackwell, Ames, IA, USA. pp 57-65 Price, D. B., Nardi, P. & Teitcher, J. (1987) Venous air embolization as a complication of pressure injection of contrast media: CT findings. Journal of Computer Assisted Tomography 11, 294-295 Riddick, L. & Brogdon, B. G. (2012) Fatal air embolism during renal dialysis. American Journal of Forensic Medicine and Pathology 33, 110-112 Rubinstein, D., Dangleis, K., Damiano, T. R. (1996) Venous air emboli identified on head and neck CT scans. Journal of Computer Assisted Tomography 20, 559-562 Sakai, O., Nakashima, N., Shinozaki, T., et al. (1998) Air bubbles in the subclavian or internal jugular veins: a common finding on contrast-enhanced CT. Neuroradiology 40, 258-260 Sodhi, K. S., Das, P. J. & Malhotra, P. (2012) Venous air embolism after intravenous contrast administration for computed tomography. The Journal of Emergency Medicine 42, 450-451 Sopena-Falco, J., Poch-Vall, N., Brullet, E., et al. (2013) Fatal massive air embolism following diagnostic colonoscopy. Endoscopy 45(Suppl 2), E91 Thayer, G. W., Carrig, C. B., Evans, A. T. (1980) Fatal venous air embolism associated with pneumocystography in a cat. Journal of American Veterinary Medical Association 176, 643-645 Woodring, J. H. & Fried, A. M. (1988) Nonfatal venous air embolism after contrastenhanced CT. Radiology 167, 405-407
© 2014 British Small Animal Veterinary Association
423
