J Neurosurg 74:1018-1020, 1991
Transesophageal echocardiography: a simple method for monitoring the patency of ventriculoatrial shunts Technical note Luc CALLIAUW,M.D., PH.D., JOHAN VANDENBOGAERDE,M.D., OKITO KALALA,M.D., JACQUES CAEMAERT,M.D., PH.D., Fa~D~alc MARTENS,M.D., AND TOM VANDEKERCKHOVE,M.D. Departments of Neurosurgery and Cardiology, UniversityHospital, Ghent, Belgium u, A new method for evaluating the patency of a ventficuloatrial shunt is described, and early experience with it is reported. Transesophageal echocardiography can demonstrate a cerebrospinal fluid leak in the fight atrium through the atrial tip of a shunting device. This capability was an incidental discovery, and since then the accuracy of the technique in evaluating the pateney of a ventriculoatrial shunt has been prospectively studied in 20 observations of 16 patients. The method proved to be accurate in 90% to 100% of cases. It is concluded that transesophageal echocardiograph:y offers a rapid and accurate assessment of ventficuloatrial shunt function, is well tolerated, and is easy to perform. KEY WORDS
"
ventriculoatrial shunt
N shunt-dependent hydrocephalic patients, the neurosurgeon often has to confirm the patency of the shunt. Monitoring of shunt function in the outpatient clinic is usually based on clinical examination and palpation of the pump. Unfortunately, the assessment is not always easy and often "pumping" is unreliable? For ventriculoatrial (VA) shunts, a more sophisticated examination, such as the injection of contrast medium z6 or radionuclides, s is used; however, repeated puncture of a shunting device can damage the valve. Shunt patency can also be evaluated by injection of radioactive isotopes into the ventricles. 37 The method of bladder scanning after introduction of ~3~I-labeled hippurate into the ventricles has often been used. This method is easy to perform in neonates but requires the creation of burr holes in adult patients. It is certainly not a noninvasive method. With the transesophageal echocardiography approach, modern ultrasound technology has provided us with a new and superior method for the dynamic imaging of dorsal cardiac structures such as the right atrium and the superior vena cava in adults, w The leakage of cerebrospinal fluid (CSF) out of the atrial tip of a VA shunt was demonstrated by one of us (J.V.) in an adult patient in whom transesophageal echocardiog-
I
~0"~8
9 shunt monitoring
9 echocardiography
raphy was performed to assess mitral valve disease. The leak presented as scattered echos around the atrial tip, which increased at each inspiration and became even more pronounced on performing a Valsalva maneuver. Impressed by the clearness of the images and the possible clinical implications of this novel technique, we performed a prospective and partly blinded study (the person conducting the study was unaware of the clinical and investigational data of the patient) to assess the accuracy of transesophageal echocardiography in examining the potency of VA shunts. This report describes the method and the initial results. Technique
Transesophageal echocardiography was performed on 20 occasions in 16 consecutive patients (nine women and seven men with a mean age of 46.9 years (range 16 to 77 years)). These patients were suspected of having an obstructed VA shunt or had placement of a new shunt. The person conducting the study was unaware of the clinical or investigational data of the patient. A commercially available 6-MHz anular array probe was used, mounted on the tip of a gastroscope 16 mm thick. The probe was connected to a commerZ Neurosurg. / Volume 74~June, 1991
T r a n s e s o p h a g e a l e c h o c a r d i o g r a p h y for s h u n t p a t e n c y out neurological complaints. Five years later he consulted a cardiologist for evaluation of cardiac symptoms. Transesophageal echocardiography showed the tip of the catheter in the right atrium and at each inspiration a scattering appeared around the tip of the catheter corresponding to a release of CSF in the atrium. This patient was obviously shunt-dependent and the device was patent. Case 2
FIG. 1. Ultrasound image obtained with a 6-MHz transesophageal probe in a patient with a ventriculoatrial shunt. The atrial tip of the shunt is correctly positioned in the middle of the right atrium. A cloud of scattered echoes emerges from the tip into the right atrium (RA) and represents the mixing of cerebrospinal fluid with blood in the right atrium. The timing of respirations can be derived from the motion of the septum primum at the level of the fossa ovalis.
This 26-year-old man was referred to us with the clinical picture of a syringomyelia. This condition was confirmed by magnetic resonance imaging, which also showed a Chiari II malformation and severe hydrocephalus. A VA shunt was placed before treatment of the syringomyelia and the Chiari malformation. Several weeks later, we requested confirmation of the patency of the shunt because the patient was still complaining of headache. With the aid of transesophageal echocardiography we were able to see the tip of the catheter slipping intermittently through the tricuspid valve. Pumping the shunting device caused a scattering of CSF to appear mostly in the right ventricle. Thus, although the system was patent, it was working inadequately because the atrial catheter was too long.
Summary of Cases
cially available echo-Doppler instrument.* Only topical pharyngeal anesthesia with a Xylocaine (lidocaine) 10% spray was used as premedication. The course of the VA shunt was imaged from the superior vena cava to its tip in the fight atrium. Initially, the correct position of the shunt in the atrium was confirmed or the incorrect position through the tricuspid valve in the fight ventricle was identified. Subsequently, the system was monitored for eventual spontaneous leakage of CSF. An example of spontaneous leakage is shown in Fig. 1. If no spontaneous leakage was demonstrated, the patient was asked to cough or squeeze (Valsalva maneuver) after which another echogram was obtained and evaluated again for CSF leak. Finally, pressure was applied to the pump in cases where there was no evidence of spontaneous or Valsalva maneuver-induced leakage. Although the studies were not timed exactly, no examination lasted more than 5 minutes. Illustrative Case Reports
The information obtained by this simple method in 16 patients is illustrated by the following two cases. Case 1
This 55-year-old man was operated on for a benign tumor in the posterior fossa. Postoperatively, he needed placement of a VA shunt. He recovered and was with* Echo-Doppler instrument manufactured by Vingmed, Horton, Norway. J. Neurosurg. / Volume 74~June, 1991
The results in this series and the comparison between echographically established shunt patency and the ultimate clinical estimation of patency in these 16 patients (20 observations) can be summarized as follows. In 14 routine observations, a good clinical evaluation correlated with a good position of the catheter and patency of the device on transesophageal echocardiography. In three observations in which the patient's clinical condition was poor, transesophageal echocardiography showed that the catheter was correctly placed but that the device was not patent. At surgery in these cases the drain proved to be obstructed. In two instances the patients were clinically unchanged after a shunting procedure, and on transesophageal echocardiography the device proved not to be patent in both cases. In one of these patients the position of the catheter was incorrect; this patient was operated on and the drain proved to be obstructed. The other patient died from a nonsurgical problem before reoperation could be performed. One patient (Case 2) had a catheter that slipped intermittently through the tricuspid valve; the catheter was patent but too long. It is obvious that the ultimate decision to evaluate the patency of a shunt surgically should be based on clinical criteria. Furthermore, in the absence of a noninvasive "gold standard," the method described here could not be compared with other techniques in cases of shunt patency. Thus, the shunt was considered to be patent if the clinical course and the investigational data did not suggest obstruction of CSF drainage, and was 1019
L. Calliauw, et al. considered to be obstructed if at surgical replacement of the device obstruction was present. Based on the data in this series, the value of transesophageal echocardiography in examining the patency of a VA shunt can be described in the following terms. If patients with unchanged clinical symptoms are considered as having a poor outcome, then the accuracy level (number of true-positive and true-negative tests) was 20 (100%) of the 20 tests performed. If, however, patients with unchanged clinical symptoms are considered as having a good outcome, then the accuracy level was 18 (90%) of the 20 tests performed. Furthermore, an incorrect position of the catheter with prolapse in the fight ventricle was evidenced in two patients. Discussion
Although our experience is thus far limited to 20 observations in 16 patients, we feel justified in claiming that it is possible to assess correctly, by means of transesophageal echocardiography, both the position and the patency of a VA shunt. Although for some awake patients the method can be inconvenient, it is well tolerated by most. In fact, as air insuffiation and suction are not used, most patients tolerate it better than a gastroduodenoscopy. The complication rate of transesophageal echocardiography is extremely low and can be expected to be lower than the 8.1% rate of upper endoscopy.~ We hope to have fewer complications due to the fact that we use no air insuffiation, suction, sclerotherapy, or biopsy. It should be stressed, however, that the probe should be inserted with gentle force by an experienced surgeon and that, in case of resistance, the procedure should be stopped. Pharyngeal and esophageal pathology, such as tumors, diverticula, and hepatic cirrhosis with esophageal varices, are absolute contraindications to the use of this technique. A major limitation to the method is its inconvenience in children. In fact, although pediatric probes exist, they are not used in children below a body weight of 10 kg; furthermore, general anesthesia is needed. In view of these limitations, we perform the procedure only in
1020
patients aged 15 years or over. Another problem is that this method only enables one to check shunt potency; it cannot localize an obstruction or quantify the rate of flow? Ventriculoatrial shunting is increasingly being abandoned in Belgium in favor of ventriculoperitoneal shunting; however, both techniques have advantages and disadvantages. Now that the intra- or postoperative monitoring of a VA device can be performed so easily with transesophageal echocardiography, we are again attracted to this type of shunt for adult patients. References
1. Chapman J, Vandenbogaerde J, Everaert J, et al: The initial clinical application of a transesophageal system with pulsed Doppler, continuous wave Doppler and color flow imaging based on an annular array technology. Int J Cardiol Imag 5:9-16, 1989 2. Dewey RC, Kosnik EJ, Sayers MP: A simple test of shunt function: the shuntgram. Technical note. J Neurosurg 44: 121-126, 1976 3. Di Chiro G, Grove AS Jr: Evaluation of surgical and spontaneous cerebrospinal fluid shunts by isotope scanning. J Neurosurg 24:743-748, 1966 4. Osaka K, Yamasaki S, Hirayama A, el al: Correlation of the response of the flushing device to compression with the clinical picture in the evaluation of the functional status of the shunting system. Childs Brain 3:25-30, 1977 5. Pendleton BD, Pollay M, Roberts PA, et al: In vitro evaluation of CSF shunt function by radionuclides. Childs Nerv Syst 1:152-157, 1985 6. Savoiardo M, Solero CL, Passerini A, et al: Determination of cerebrospinal fluid shunt function with water-soluble contrast medium. J Neurosurg 49:398-407, 1978 7. Seward JB, Khandheria BK, Abel MD, et al: Transesophageal echocardiography: technique, anatomic correlations, implementation, and clinical applications. Mayo Clin Proc 63:649-680, 1988
Manuscript received June 1, 1990. Accepted in final form October 26, 1990. Address reprint requests to: Luc Calliauw, M.D., Ph.D., Department of Neurosurgery, De Pintelaan 185, 9000 Ghent, Belgium.
J. Neurosurg. / Volume 74~June, 1991
Transesophageal echocardiography: a simple method for monitoring the patency of ventriculoatrial shunts Technical note Luc CALLIAUW,M.D., PH.D., JOHAN VANDENBOGAERDE,M.D., OKITO KALALA,M.D., JACQUES CAEMAERT,M.D., PH.D., Fa~D~alc MARTENS,M.D., AND TOM VANDEKERCKHOVE,M.D. Departments of Neurosurgery and Cardiology, UniversityHospital, Ghent, Belgium u, A new method for evaluating the patency of a ventficuloatrial shunt is described, and early experience with it is reported. Transesophageal echocardiography can demonstrate a cerebrospinal fluid leak in the fight atrium through the atrial tip of a shunting device. This capability was an incidental discovery, and since then the accuracy of the technique in evaluating the pateney of a ventriculoatrial shunt has been prospectively studied in 20 observations of 16 patients. The method proved to be accurate in 90% to 100% of cases. It is concluded that transesophageal echocardiograph:y offers a rapid and accurate assessment of ventficuloatrial shunt function, is well tolerated, and is easy to perform. KEY WORDS
"
ventriculoatrial shunt
N shunt-dependent hydrocephalic patients, the neurosurgeon often has to confirm the patency of the shunt. Monitoring of shunt function in the outpatient clinic is usually based on clinical examination and palpation of the pump. Unfortunately, the assessment is not always easy and often "pumping" is unreliable? For ventriculoatrial (VA) shunts, a more sophisticated examination, such as the injection of contrast medium z6 or radionuclides, s is used; however, repeated puncture of a shunting device can damage the valve. Shunt patency can also be evaluated by injection of radioactive isotopes into the ventricles. 37 The method of bladder scanning after introduction of ~3~I-labeled hippurate into the ventricles has often been used. This method is easy to perform in neonates but requires the creation of burr holes in adult patients. It is certainly not a noninvasive method. With the transesophageal echocardiography approach, modern ultrasound technology has provided us with a new and superior method for the dynamic imaging of dorsal cardiac structures such as the right atrium and the superior vena cava in adults, w The leakage of cerebrospinal fluid (CSF) out of the atrial tip of a VA shunt was demonstrated by one of us (J.V.) in an adult patient in whom transesophageal echocardiog-
I
~0"~8
9 shunt monitoring
9 echocardiography
raphy was performed to assess mitral valve disease. The leak presented as scattered echos around the atrial tip, which increased at each inspiration and became even more pronounced on performing a Valsalva maneuver. Impressed by the clearness of the images and the possible clinical implications of this novel technique, we performed a prospective and partly blinded study (the person conducting the study was unaware of the clinical and investigational data of the patient) to assess the accuracy of transesophageal echocardiography in examining the potency of VA shunts. This report describes the method and the initial results. Technique
Transesophageal echocardiography was performed on 20 occasions in 16 consecutive patients (nine women and seven men with a mean age of 46.9 years (range 16 to 77 years)). These patients were suspected of having an obstructed VA shunt or had placement of a new shunt. The person conducting the study was unaware of the clinical or investigational data of the patient. A commercially available 6-MHz anular array probe was used, mounted on the tip of a gastroscope 16 mm thick. The probe was connected to a commerZ Neurosurg. / Volume 74~June, 1991
T r a n s e s o p h a g e a l e c h o c a r d i o g r a p h y for s h u n t p a t e n c y out neurological complaints. Five years later he consulted a cardiologist for evaluation of cardiac symptoms. Transesophageal echocardiography showed the tip of the catheter in the right atrium and at each inspiration a scattering appeared around the tip of the catheter corresponding to a release of CSF in the atrium. This patient was obviously shunt-dependent and the device was patent. Case 2
FIG. 1. Ultrasound image obtained with a 6-MHz transesophageal probe in a patient with a ventriculoatrial shunt. The atrial tip of the shunt is correctly positioned in the middle of the right atrium. A cloud of scattered echoes emerges from the tip into the right atrium (RA) and represents the mixing of cerebrospinal fluid with blood in the right atrium. The timing of respirations can be derived from the motion of the septum primum at the level of the fossa ovalis.
This 26-year-old man was referred to us with the clinical picture of a syringomyelia. This condition was confirmed by magnetic resonance imaging, which also showed a Chiari II malformation and severe hydrocephalus. A VA shunt was placed before treatment of the syringomyelia and the Chiari malformation. Several weeks later, we requested confirmation of the patency of the shunt because the patient was still complaining of headache. With the aid of transesophageal echocardiography we were able to see the tip of the catheter slipping intermittently through the tricuspid valve. Pumping the shunting device caused a scattering of CSF to appear mostly in the right ventricle. Thus, although the system was patent, it was working inadequately because the atrial catheter was too long.
Summary of Cases
cially available echo-Doppler instrument.* Only topical pharyngeal anesthesia with a Xylocaine (lidocaine) 10% spray was used as premedication. The course of the VA shunt was imaged from the superior vena cava to its tip in the fight atrium. Initially, the correct position of the shunt in the atrium was confirmed or the incorrect position through the tricuspid valve in the fight ventricle was identified. Subsequently, the system was monitored for eventual spontaneous leakage of CSF. An example of spontaneous leakage is shown in Fig. 1. If no spontaneous leakage was demonstrated, the patient was asked to cough or squeeze (Valsalva maneuver) after which another echogram was obtained and evaluated again for CSF leak. Finally, pressure was applied to the pump in cases where there was no evidence of spontaneous or Valsalva maneuver-induced leakage. Although the studies were not timed exactly, no examination lasted more than 5 minutes. Illustrative Case Reports
The information obtained by this simple method in 16 patients is illustrated by the following two cases. Case 1
This 55-year-old man was operated on for a benign tumor in the posterior fossa. Postoperatively, he needed placement of a VA shunt. He recovered and was with* Echo-Doppler instrument manufactured by Vingmed, Horton, Norway. J. Neurosurg. / Volume 74~June, 1991
The results in this series and the comparison between echographically established shunt patency and the ultimate clinical estimation of patency in these 16 patients (20 observations) can be summarized as follows. In 14 routine observations, a good clinical evaluation correlated with a good position of the catheter and patency of the device on transesophageal echocardiography. In three observations in which the patient's clinical condition was poor, transesophageal echocardiography showed that the catheter was correctly placed but that the device was not patent. At surgery in these cases the drain proved to be obstructed. In two instances the patients were clinically unchanged after a shunting procedure, and on transesophageal echocardiography the device proved not to be patent in both cases. In one of these patients the position of the catheter was incorrect; this patient was operated on and the drain proved to be obstructed. The other patient died from a nonsurgical problem before reoperation could be performed. One patient (Case 2) had a catheter that slipped intermittently through the tricuspid valve; the catheter was patent but too long. It is obvious that the ultimate decision to evaluate the patency of a shunt surgically should be based on clinical criteria. Furthermore, in the absence of a noninvasive "gold standard," the method described here could not be compared with other techniques in cases of shunt patency. Thus, the shunt was considered to be patent if the clinical course and the investigational data did not suggest obstruction of CSF drainage, and was 1019
L. Calliauw, et al. considered to be obstructed if at surgical replacement of the device obstruction was present. Based on the data in this series, the value of transesophageal echocardiography in examining the patency of a VA shunt can be described in the following terms. If patients with unchanged clinical symptoms are considered as having a poor outcome, then the accuracy level (number of true-positive and true-negative tests) was 20 (100%) of the 20 tests performed. If, however, patients with unchanged clinical symptoms are considered as having a good outcome, then the accuracy level was 18 (90%) of the 20 tests performed. Furthermore, an incorrect position of the catheter with prolapse in the fight ventricle was evidenced in two patients. Discussion
Although our experience is thus far limited to 20 observations in 16 patients, we feel justified in claiming that it is possible to assess correctly, by means of transesophageal echocardiography, both the position and the patency of a VA shunt. Although for some awake patients the method can be inconvenient, it is well tolerated by most. In fact, as air insuffiation and suction are not used, most patients tolerate it better than a gastroduodenoscopy. The complication rate of transesophageal echocardiography is extremely low and can be expected to be lower than the 8.1% rate of upper endoscopy.~ We hope to have fewer complications due to the fact that we use no air insuffiation, suction, sclerotherapy, or biopsy. It should be stressed, however, that the probe should be inserted with gentle force by an experienced surgeon and that, in case of resistance, the procedure should be stopped. Pharyngeal and esophageal pathology, such as tumors, diverticula, and hepatic cirrhosis with esophageal varices, are absolute contraindications to the use of this technique. A major limitation to the method is its inconvenience in children. In fact, although pediatric probes exist, they are not used in children below a body weight of 10 kg; furthermore, general anesthesia is needed. In view of these limitations, we perform the procedure only in
1020
patients aged 15 years or over. Another problem is that this method only enables one to check shunt potency; it cannot localize an obstruction or quantify the rate of flow? Ventriculoatrial shunting is increasingly being abandoned in Belgium in favor of ventriculoperitoneal shunting; however, both techniques have advantages and disadvantages. Now that the intra- or postoperative monitoring of a VA device can be performed so easily with transesophageal echocardiography, we are again attracted to this type of shunt for adult patients. References
1. Chapman J, Vandenbogaerde J, Everaert J, et al: The initial clinical application of a transesophageal system with pulsed Doppler, continuous wave Doppler and color flow imaging based on an annular array technology. Int J Cardiol Imag 5:9-16, 1989 2. Dewey RC, Kosnik EJ, Sayers MP: A simple test of shunt function: the shuntgram. Technical note. J Neurosurg 44: 121-126, 1976 3. Di Chiro G, Grove AS Jr: Evaluation of surgical and spontaneous cerebrospinal fluid shunts by isotope scanning. J Neurosurg 24:743-748, 1966 4. Osaka K, Yamasaki S, Hirayama A, el al: Correlation of the response of the flushing device to compression with the clinical picture in the evaluation of the functional status of the shunting system. Childs Brain 3:25-30, 1977 5. Pendleton BD, Pollay M, Roberts PA, et al: In vitro evaluation of CSF shunt function by radionuclides. Childs Nerv Syst 1:152-157, 1985 6. Savoiardo M, Solero CL, Passerini A, et al: Determination of cerebrospinal fluid shunt function with water-soluble contrast medium. J Neurosurg 49:398-407, 1978 7. Seward JB, Khandheria BK, Abel MD, et al: Transesophageal echocardiography: technique, anatomic correlations, implementation, and clinical applications. Mayo Clin Proc 63:649-680, 1988
Manuscript received June 1, 1990. Accepted in final form October 26, 1990. Address reprint requests to: Luc Calliauw, M.D., Ph.D., Department of Neurosurgery, De Pintelaan 185, 9000 Ghent, Belgium.
J. Neurosurg. / Volume 74~June, 1991
