Intraoperative Ultrasound Guidance for the Placement of Permanent Ventricular Cerebrospinal Fluid Shunt Catheters: A Single-Center Historical Cohort Study

Intraoperative Ultrasound Guidance for the Placement of Permanent Ventricular Cerebrospinal Fluid Shunt Catheters: A Single-Center Historical Cohort Study R. Webster Crowley1, Aaron S. Dumont2, Ashok R. Asthagiri3, James C. Torner4, Ricky Medel1, John A. Jane, Jr.1, John A. Jane1, Neal F. Kassell1

Key words Cerebrospinal fluid shunts - Hydrocephalus - Image-guided surgery -

Abbreviations and Acronyms CSF: Cerebrospinal fluid CT: Computed tomography NPH: Normal pressure hydrocephalus US: Ultrasound VCR: Ventricular catheter revision From the 1Department of Neurological Surgery, University of Virginia School of Medicine, Charlottesville, Virginia; 2Department of Neurological Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania; 3Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland; and 4Department of Epidemiology, University of Iowa College of Public Health, Iowa City, Iowa, USA To whom correspondence should be addressed: R. Webster Crowley, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2014) 81, 2:397-403. http://dx.doi.org/10.1016/j.wneu.2013.01.039 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2014 Elsevier Inc. All rights reserved.

- OBJECTIVE:

Despite the frequency with which ventriculoperitoneal shunts are placed, ventricular catheter revision rates remains as high as 30%e40% at 1 year. Many neurosurgeons place ventricular catheters “blindly” depending on anatomical landmarks and personal experience. To determine whether intraoperative ultrasonography is beneficial for ventricular catheter placement, we performed a historical cohort study comparing shunts placed with intraoperative ultrasound (US) guidance to those placed blindly.

- METHODS:

We reviewed all shunts placed by the Department of Neurosurgery at the University of Virginia from January 2005 to January 2007. During that time 211 patients underwent 242 shunts, with US use determined by surgeon’s preference. Ninety-two shunts were placed by the use of US guidance, and 150 were placed without US. Adults received 176 shunts, 56 with US. Children received 66 shunts, 36 with US. Mean follow-up was 21.6 months. The primary end points examined were shunt revision, ventricular catheter revision (VCR), and acute VCR (revision within 1 week for an improperly-placed catheter).

- RESULTS:

The use of US was associated with a statistically significant decrease in shunt revisions (odds ratio 0.492; 95% confidence interval 0.253e0.958). Of the shunts placed with US guidance, 21.7% required revision, compared with 29.3% without US. VCRs and acute VCRs occurred in 9.8% and 2.2%, respectively, for US shunts, compared with 14% and 5.3% without US. Pediatric revision rates were 30.6% with US versus 53.3% without, whereas adult rates were 16.1% and 23.3%, respectively. The benefit of US was more profound for occipital shunts.

- CONCLUSIONS:

The use of US for the placement of permanent cerebrospinal fluid shunt catheters is associated with a decreased risk of shunt revision.

INTRODUCTION Ventriculoperitoneal shunts are among the most common procedures performed by neurosurgeons, yet revision rates have been reported as high as 30%e40% within the first year after placement (4, 6, 11). Despite the fact that the improper passage of a ventricular catheter may have catastrophic consequences, common practice remains the use of “blind” catheter placement. With blind catheter placement, the entry site and trajectory is selected with only anatomical landmarks, preoperative imaging, and personal experience as guides. In addition to the potential for acute neurological morbidity associated with a misplaced ventricular catheter, catheters that are suboptimally placed may be at a greater risk for eventual

proximal catheter obstruction, necessitating shunt revision. Real-time intraoperative ultrasound (US) guidance increasingly is being used as a surgical adjunct to provide immediate feedback to the surgeon for a variety of neurosurgical procedures, including the placement of ventricular catheters (3, 15, 17, 18). US-guided ventricular catheter placement was first described in infants through open fontanelles. Subsequently, with the development of smaller US probes, this technology was applied to ventricular catheter placement in older children and adults via a burr hole (17, 18). To determine whether real-time intraoperative ultrasonography improves ventricular catheter

WORLD NEUROSURGERY 81 [2]: 397-403, FEBRUARY 2014

placement and decreases rates of shunt revision, we performed a historical cohort study comparing revision rates in patients with permanent ventricular catheters placed with intraoperative US guidance versus those placed “blindly.”

MATERIALS AND METHODS Patient Selection All patients who underwent placement of a new, lateral ventricle shunt catheter during a 25-month period from January 2005 through January 2007 at the University of Virginia were included in this institutional review board
approved study, regardless of

www.WORLDNEUROSURGERY.org

397

HYDROCEPHALUS

PEER-REVIEW REPORTS

PEER-REVIEW REPORTS R. WEBSTER CROWLEY ET AL.

HYDROCEPHALUS

whether it was an initial shunt placement or revision of a pre-existing shunt. Only shunts placed via a frontal burr hole (Kocher’s point) or an occipital burr hole (Frazier’s point) were included in this study. Fourth ventricular shunts and shunts siphoning the subdural space or intracranial cysts were excluded. The primary end points of interest were total revisions, defined as revisions for any reason (e.g., obstruction of the valve, ventricular catheter or distal catheter, or infection); ventricular catheter revisions; and acute ventricular catheter revisions, defined as those catheters that required revision within 7 days of the initial surgery because of improper placement. Follow-up was determined for each individual shunt catheter as either: (1) the date of ventricular catheter revision; (2) the date of the most recent clinic visit; (3) the date of phone interview for those patients who have not been seen recently in clinic; or (4) the date of death. Operative Technique The technique used for placing a ventricular catheter through an enlarged burr hole with the aid of real-time US guidance has been previously described (18). In brief, the patient is positioned and draped in typical sterile fashion for shunt placement. A portable US scanner (ProSound Alpha 7, Aloka, Inc., Wallingford, Connecticut, USA) with a sterilized bayonet-shaped transducer (Multifrequency Burr-Hole Transducer; Aloka, Inc., Wallingford, Connecticut, USA) is used (Figure 1). Using a standard adult perforator (14 mm; Acra-Cut, Inc., Acton, Massachusetts, USA), a burr hole is made. At

US GUIDANCE TO PLACE VENTRICULAR CSF SHUNT CATHETERS

this time, the US is used to assess the adequacy of the location of the burr hole and determine which margin would be preferred for catheter entry and provide an ideal trajectory to the lateral ventricle. The burr hole is then widened in this direction to a width of approximately 2 cm. The dura over the burr hole is opened and a corticotomy performed to allow catheter passage. Once the planned site of the catheter insertion is prepared, the US transducer is gently placed in the burr hole. For a frontal shunt, the transducer is rotated orthogonally until a coronal view is obtained. The US transducer can then be angled in an anteroposterior direction until all pertinent anatomy is visualized (bilateral frontal horns and foramena of Monro, falx, third ventricle, and choroid plexus). The optimal target is approximately 1e2 cm from the foramen of Monro in the ipsilateral lateral ventricle, ensuring that the tip does not contact the choroid plexus. It is therefore critical to visualize the foramen of Monro and the choroid plexus prior to catheter passage. For an occipital shunt, the transducer is manipulated until the trigone is appreciated. With the same optimal target used for the catheter tip as frontal shunts, the catheter is visualized entering the trigone and passed 2e3 cm further under continuous US guidance without the stylet to place it in the frontal horn. Some US scanners have a particularly helpful function that superimposes a dashed line onto the US image, signifying the real-time expected catheter trajectory as determined by the location and direction of the transducer head.

Once the appropriate trajectory is chosen, the US transducer is held in place by one surgeon, while an assistant surgeon passes the ventricular catheter along the groove of the transducer under real-time US guidance. In addition to the tactile feedback of entering the ventricle, direct visualization of the catheter entering the ventricle is obtained. Once the stylet is removed the perforated portion of the catheter can be seen within the ventricle, an advantage that usually ensures that the ventricular catheter is sufficiently in place. Catheters deemed to be suboptimally placed were repositioned when necessary. The remainder of the shunt procedure was performed with standard techniques. The majority of the shunts were constructed of a ventricular catheter and a unitized inline valve and distal catheter. This occasionally varied, however, particularly in the case of revision surgeries where the original system was kept in place when possible. We do not routinely perform a postassembly US check to see whether the ventricular catheter moved during the course of the shunt connection. Instead, we rely on length markings on the ventricular catheter and, when possible, flow through the distal catheter to ensure the catheter has not been moved and is continuing to provide a conduit for cerebrospinal fluid (CSF). Statistical Analysis Proportions, means, and ranges were calculated to describe the population. Univariate comparisons of US guidance or not were performed with t-tests for continuous variables and c2 tests for proportions. An alpha level of 0.05 was used for significance. The comparison of the primary end points was done with the use of multiple logistic regression with the PROC LOGIST procedure of SAS, Version 9.2 (SAS Institute, Cary, North Carolina, USA). Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to determine significance. RESULTS

Figure 1. Photographs of (A) portable ultrasound scanner and (B) bayonet-shaped transducer. The groove on the transducer head accommodates the ventricular catheter during placement.

398

www.SCIENCEDIRECT.com

Total Population From January 2005 to January 2007, 211 patients underwent a total of 242 shunt procedures in which a new ventricular catheter was placed. Of these shunts, 92 (38%) were placed with the aid of intraoperative US guidance, whereas 150 (62%) were placed “blindly.” The mean age of all

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2013.01.039

PEER-REVIEW REPORTS US GUIDANCE TO PLACE VENTRICULAR CSF SHUNT CATHETERS

patients was 44.9 years. Mean follow-up was 21.6 months (range, 1 day to 51 months). Table 1 demonstrates the variety of diagnoses requiring shunt placement. Ventricular catheters were placed via a frontal burr hole (n ¼ 183, 75.6%) or via an occipital burr hole (n ¼ 59, 24.4%). Occipital shunts constituted a greater proportion of the US-assisted surgeries (38% occipital, 62% frontal) compared with those performed without US (16% occipital, 84% frontal; P < 0.0001). Similarly, the US was more likely to be used with revision surgeries, as 33.7% of USguided cases were revisions versus 23.3% of non-US cases. Two-hundred thirtyseven patients underwent ventriculoperitoneal shunts, whereas five shunts involved placement of the distal catheter elsewhere (two ventriculoatrial, two ventriculocholecystic, one ventriculopleural; P ¼ 0.04; Table 2). Revisions were eventually required in 26.5% (64/242) of all cases. For shunts placed with US guidance, 21.7% required revision compared with 29.3% of those placed without US. Decreased rates also were associated with the use of the US for both ventricular catheter revisions and acute

ventricular catheter revisions. Nine (9/92, 9.8%) ventricular shunt catheters placed with US guidance eventually required revision of the ventricular catheter, compared with 14% (21/150) for those placed without US. For the ventricular catheters that required acute revision (within 7 days) because of an improper catheter location, 2.2% (2/92) of the US group required an acute revision, a percentage that was less than one half of that observed the non-US group (5.3%, 8/150; Table 3). When comparing frontal and occipital shunts, the benefit of the US was more pronounced for those catheters placed via an occipital burr hole. The US was associated with a difference of 2.4% in ventricular catheter revision rates for those placed via a frontal burr hole, whereas it was associated with a difference of 16.4% for those placed via an occipital burr hole (Table 4). The average time to revision for any reason was 191.1 days (range, 1e1160 days), a finding that was not significantly different with or without US (Table 5). However, when looking at the time to revision for an obstructed or improperly placed ventricular catheter, shunts placed with US guidance

Table 1. Number of Shunts Placed with and without Ultrasound, by Patient Diagnosis for the Total (P < 0.0001), Adult (P < 0.0001), and Pediatric (P ¼ 0.054) Populations Patient Diagnosis

US

Non-US

US Adults

Non-US Adults

US Pediatrics

Non-US Pediatrics

Normal pressure hydrocephalus

11

72

11

72

0

0

5

6

5

5

0

1

Pseudotumor cerebri Acqueductal stenosis

4

13

1

1

3

12

Posterior fossa lesion

10

10

5

8

5

2

Supratentorial lesion

8

9

8

8

0

1

Myelomeningocele

10

7

1

2

9

5

Intraventricular hemorrhage

10

5

3

0

7

5

Posttrauma

8

6

4

6

4

0

Subarachnoid hemorrhage

5

4

5

4

0

0

CSF leak

4

3

2

3

2

0

Intraventricular lesion

4

2

3

2

1

0

Dandy-Walker malformation

2

2

0

0

2

2

Postmeningitis

2

3

2

3

0

0

Other

9

8

6

6

3

2

Total

92

150

56

120

36

30

US, ultrasound; CSF, cerebrospinal fluid.

WORLD NEUROSURGERY 81 [2]: 397-403, FEBRUARY 2014

had a substantially longer time-to-revision than those placed without the aid of US (272.3 vs. 91.3 days). When we excluded all shunts that were acutely revised because of an improperly placed ventricular catheter, we detected a substantial improvement in time-to-revision persisted for those shunts placed with US guidance (349.6 vs. 146.2 days). When multivariate logistic regression analysis was performed, the use of intraoperative US guidance was associated with a decreased risk of revision compared with those shunts placed without US guidance (OR 0.492; 95% CI 0.253e0.958). Adult Population The adult population consisted of 156 patients who underwent a total of 176 shunts (Table 1). Of these shunts, 56 (31.8%) were placed with the use of the US, whereas 120 (68.2%) were placed without US. The mean age in this population was 60.1 years (range, 19e88 years). The mean follow-up was 22.2 months (range, 1 day to 51 months). The surgery was an initial shunt placement in 71.4% and 79.2% of cases with and without US, respectively. Ventricular catheters were placed via a frontal burr hole in 58.9% of US cases compared with 88.3% of the non-US cases. Revisions attributable to all causes were eventually required in 16.1% (9/56) of cases that used US guidance compared with 23.3% (28/120) of cases without US. When we examined ventricular catheter revisions in adults, the “blind” cases were more than twice as likely to require revisions than those placed with the US (11.7% [14/120] vs. 5.4% [3/56]). In addition, acute catheter revisions were nearly four times as likely to be performed for non-US cases because only 1 of 56 (1.8%) cases in which surgeons used US required acute ventricular catheter revision, whereas 8 of 120 (6.7%) of non-U/ S cases had to be revised because of poor placement. Multivariate logistic regression analysis showed a trend towards decreased risk of shunt revision with the use of US guidance; however, this finding did not reach statistical significance (OR 0.531; 95% CI 0.212e1.328). Pediatric Population For the pediatric population, 55 patients underwent a total of 66 shunts (Table 1). Of these shunts, 36 (54.5%) were placed using intraoperative US guidance, whereas 30 (45.5) were placed without US. The

www.WORLDNEUROSURGERY.org

399

HYDROCEPHALUS

R. WEBSTER CROWLEY ET AL.

PEER-REVIEW REPORTS R. WEBSTER CROWLEY ET AL.

US GUIDANCE TO PLACE VENTRICULAR CSF SHUNT CATHETERS

HYDROCEPHALUS

Table 2. Shunt Characteristics for the Total Population, Adult Population, and Pediatric Population Total Population

Adult

Ultrasound No Mean age, years Initial surgery

Ultrasound

Yes

52

Pediatric

33.2

P Value

No

< 0.0001

64.3

Ultrasound

Yes

P Value

31.2

0.0001 0.258

No

Yes

P Value

3

5.3

ns

76.7% 66.3%

0.079

79.2% 71.4%

66.7% 58.3%

Frontal

84.0% 62.0%

0.0001

88.3% 58.9% < 0.0001 66.7% 66.7%

Occipital

16.0% 38.0%

0.487

Location of vent

11.7% 41.1%

ns

33.3% 33.3%

Type of Surgery 0.0%

2.2%

V-Pleural

0.0%

1.1%

0.0%

2.2%

VA VP

0.04

100.0% 94.6%

0.0%

1.8%

0.0%

0.0%

0.0%

3.6%

100.0% 94.6%

0.038

2.8%

Occipital US

35

3 (8.6)

Occipital non-US

24

6 (25)

Frontal US

57

6 (10.5)

Frontal nonUS

126

15 (12.9)

0.423

2.8% 0.0% 100.0%

94.4%

V-GB, ventriculo-gall bladder; VA, ventriculoatrial; VP, ventriculoperitoneal.

mean age was 4.3 years (range, 1 day to 17 years) and mean follow-up was 20.1 months (range, 1 day to 47 months). Shunt catheters were placed as part of an initial shunt surgery in 58.3% of US cases, and 66.7% of non-US cases. Catheters placed via a frontal burr hole consisted of 66.7% of cases both with and without the US. Revisions attributable to all causes were eventually required in 53.3% (16/30) of non-US cases versus 30.6% (11/36) of cases using US. Ventricular catheter revisions were also more frequent in the non-US group (23.3% vs. 16.7%). Of all the pediatric shunts placed, only one case (1.5%), a US-assisted surgery, required acute

ventricular catheter revision secondary to an improperly placed ventricular catheter, whereas none of the 30 cases performed without US required acute ventricular catheter revision. Multivariate logistic regression analysis again showed a trend towards a decreased risk of shunt revision with US guidance that did not reach statistical significance (OR 0.434; 95% CI 0.156e1.208). DISCUSSION A number of intraoperative image guidance techniques have been described for the placement of ventricular catheters,

Total Population Ultrasound

Pediatric

NonUltrasound Ultrasound

Ultrasound

No

Yes

P Value

No

Yes

P Value

No

Yes

P Value

Revision

29.3%

21.7%

0.194

53.3%

30.6%

0.061

23.3%

16.1%

0.271

VCR

14.0%

9.8%

0.334

23.3%

16.7%

0.498

11.7%

5.4%

0.187

5.3%

2.2%

0.231

0.0%

2.8%

0.358

6.7%

1.8%

0.171

www.SCIENCEDIRECT.com

Existing Literature The first description of ventricular catheter placement using real-time US was published in 1981 by Shkolnik and McLone (15). They described its use with 7 hydrocephalic pediatric patients, 6 of whom had previously undergone myelomeningocele

Adult

Ultrasound

VCR, ventricular catheter revision.

including endoscopic navigation and electromagnetic navigation (2, 7, 8, 10). Of the various techniques that exist, US is arguably one of the easiest, most inexpensive tools that can be used to obtain real-time, intraoperative visualization. In addition to its diagnostic capabilities, its utility and feasibility have been previously demonstrated for a variety of neurosurgical procedures and pathological conditions, including tumor biopsy, epilepsy surgery, cyst drainage, arteriovenous malformation surgery, distal shunt catheter placement, and in the evacuation of abscess and hematomas (1, 5, 9, 12-14, 17).

Table 5. Average Time to Revision for Shunts Placed with and without Ultrasound Guidance (P ¼ NS)

Table 3. Revision Rates for the Total Population, Pediatric Population, and Adult Population

400

Location

Ventricular Total Catheter Catheters Revisions, n (%)

US, ultrasound.

V-GB

Acute VCR

Table 4. Number of Ventricular Catheters Placed and Revised, According to Location of Burr Hole and Use of US Guidance

Time to revision (days): all causes

190.6

191.4

Time to revision (days): ventricular catheter

272.3

91.3

Time to revision (days): nonacute ventricular catheter

349.6

146.2

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2013.01.039

PEER-REVIEW REPORTS

repairs. These patients included 5 neonates, and 2 infants younger than 8 months of age, all with patent anterior fontanelles. The ventricular catheters were passed using anatomical landmarks via a twist drill hole in the posterior parietal area; however, once the catheter was in place the U/S was placed over the anterior fontanelle and was used to confirm its placement, and to guide the catheter tip to its desired location. This was thought to minimize the risk of catheter obstruction by the choroid plexus. Shortly thereafter, Rubin and Dohrmann applied this technique to the placement of ventricular catheters in adults (13). They described five cases (two ventriculoperitoneal shunts, three Ommaya reservoirs) in which they performed a craniotomy to accommodate the US probe, followed by placement of the ventricular catheter using intraoperative US guidance. Although their method requires a larger craniotomy than currently used, and the US used today are of better quality, today’s methods are nothing more than a refinement of their technique. Despite the integration of these techniques nearly 30 years ago, there have been relatively few series investigating the effectiveness of US guidance for ventricular catheter placement. In 2000, Strowitzki et al. (17) published a series of 100 neurosurgical procedures performed with the aid of intraoperative US. Using a bayonet-shaped transducer similar to ours, they described its usefulness in tapping intracranial cysts, evacuating abscesses or hematomas, biopsying intracranial tumors, and, in 46 patients, tapping the ventricular system. Of these 46 patients, they were unable to visualize the ventricles in 12 patients but were able to access the ventricles in all but one of the remainder of the patients, with an average of 1.1 passes with use of the US. Their series did not differentiate between cases involving permanent CSF diversion and temporary CSF diversion, nor did it include follow-up. Several years later, Strowitzki et al. (16) performed a matched pair analysis of 115 patients who underwent tapping of their ventricular system for temporary CSF diversion, monitoring of intracranial pressure, or permanent shunt placement. Of these patients, 48 had their procedure performed with the assistance of US guidance. No mention was made of the number of permanent shunts that were included in this

US GUIDANCE TO PLACE VENTRICULAR CSF SHUNT CATHETERS

study, and no follow-up was included. The authors did, however, observe an increase in the accuracy of the catheter tip placement for those catheters placed with US guidance, although there was no difference in the number of attempts needed to access the ventricle. In 2007, Whitehead et al. (18) published a technical note describing the procedure, which also involved enlarging the standard burr hole in order to accommodate the US probe. Although they did not compare the US group with a control group, they described 10 pediatric patients with closed fontanelles who successfully underwent placement of ventricular catheters with the assistance of U/S-guidance, none of which required more than 1 pass to the ventricle. Current Study The results of our study indicate that US guidance for permanent ventricular catheter placement is associated with a statistically significant decrease in the risk of shunt revision. In addition, although our study was not powered to detect statistically significant differences in ventricular catheter revisions and acute ventricular catheter revisions, we observed a substantially decreased rate in both of these end points for shunts placed with US guidance compared with those placed without US, a decreased rate that we believe is real and would become significant with a largeenough patient sample. The reasons for these differences are not entirely clear; however, regarding ventricular catheter revisions, we believe the shunts placed without US were more likely to abut the choroid plexus, or other periventricular structures, and were more likely to have the catheter tip terminate in the contralateral lateral ventricle or in the third ventricle. These shunts would not have been revised acutely as long as the patient was doing well clinically, and follow-up head CT demonstrated interval decrease in ventricular size. With regards to acute ventricular catheter revisions, we believe that the main reason for these occurring was improper trajectory during catheter placement. One speculation is that this may have been related to resident experience level, however this is not verifiable. The improvement in ventricular catheter revision rates conferred by US was more pronounced for occipital shunts. Although

WORLD NEUROSURGERY 81 [2]: 397-403, FEBRUARY 2014

in previous studies authors have found that occipital shunts are more likely to be revised than frontal shunts (6), the patients in our series who underwent occipital shunts with US had revision rates that were comparable with frontal shunts. With regard to the time between catheter placement and catheter revision, we found that US guidance was associated with an approximately 3-fold increase in the time to revision for the shunts that required ventricular catheter revision. This difference in time-to-revision persisted, and in fact widened to greater than 200 days, when we excluded those shunts that required revision within seven days for improperly-placed catheters. This finding suggests that the use of US not only decreases the rates of misplaced ventricular catheters but also lengthens the lifespan of those catheters that are initially placed within the ventricle. The possible reasons for this have been previously discussed, but again may be related to an increased likelihood of adjacency to the choroid plexus or placement in the third or contralateral lateral ventricles. Certainly ventricular catheter failure, particularly in pediatric patients, may result from reconstitution of the cortical mantle or withdrawal of the catheter due to growth; however, this is unlikely to affect our study populations disproportionately. Although the use of the US was based on physician preference, the decision to use US often varied from case to case among individual physicians. Cases in which difficulty was anticipated preoperatively, or where “blind” placement had previously resulted in a poorly placed catheter, commonly were inserted via the use of US (Figure 2), whereas relatively straightforward shunts for normal pressure hydrocephalus (NPH) were usually performed without US. Certainly familiarity, or lack thereof, with the US may have also impacted the decision to use this technology. Although we believe our study has important implications, it is important to recognize its shortcomings. Perhaps the most significant of these is the fact that the patient populations were not matched according to diagnosis, which resulted in several discrepancies between the two study populations. The most noticeable of these discrepancies was that observed with the diagnosis of NPH. NPH constituted

www.WORLDNEUROSURGERY.org

401

HYDROCEPHALUS

R. WEBSTER CROWLEY ET AL.

PEER-REVIEW REPORTS R. WEBSTER CROWLEY ET AL.

US GUIDANCE TO PLACE VENTRICULAR CSF SHUNT CATHETERS

HYDROCEPHALUS Figure 2. Postoperative head computed tomography scans demonstrating ventricular catheter placement in a patient with pseudotumor cerebri. (A) Improper placement of the ventricular catheter after surgery without ultrasound (US) guidance. The patient returned to the operating room the following day for placement of the ventricular catheter using US guidance. (B) Adequate placement of the catheter after the revision surgery.

34.3% of all cases; however, it was the primary diagnosis in 48% of patients who underwent shunt placement without the aid of US versus 12% of those shunts placed with US guidance. This difference was obviously even more pronounced when we examined the adult population, in which NPH was the primary diagnosis for 60% of the non-US group and 20% of the US group. Although discrepancies such as this one make our analysis more difficult, we believe that this significant difference understates the benefits that we have observed with U/S use. In other words, it is possible, and perhaps likely, that a number of patients with NPH have shunts that are nonfunctioning; however, because of the nature of the disease, they may not be recognized as such. If this were the case, it would almost certainly disproportionately affect the nonUS group, and therefore our observed failure rates for non-US shunts may be falsely low. Subsequently, the benefit conferred by the US may actually be greater than we have observed. Of course this argument points to another shortcoming of our study, which is the fact that not all patients have undergone shunt interrogations to determine whether their shunt is, in fact, working. Therefore, it is unclear whether all patients have functioning shunts or whether some have nonfunctioning shunts but have become shunt-independent. Although it is likely that

402

www.SCIENCEDIRECT.com

some patients have unrecognized shunt failure, it is our belief that aside from those cases in the NPH population, this would be unlikely to disproportionately affect either the US or non-US groups. Finally, as a retrospective study, selection bias may be present, and certain data were not available that may have contributed to our study. For instance, the number of passes taken before successfully entering the ventricle typically was not described in our operative notes, and therefore we were unable to determine whether the use of intra-operative US resulted in fewer passes. Anecdotally, however, we are convinced that the US results in fewer passes before successfully entering the ventricle. Although we believe that the intraoperative US has a definite utility in the placement of shunts, it is important to caution that the US should be used in conjunction with, but not in lieu of, a sound understanding of the anatomical landmarks typically used for ventricular catheter placement. Becoming overly reliant on intraoperative guidance, whether it is the US or other forms of neuronavigation, will not only significantly hinder resident education but can have disastrous consequences should the equipment fail, and even in instances when it does not. Figure 3 demonstrates one such case, in which intraoperative US was used to place an occipital ventricular catheter. Although the

Figure 3. Postoperative head computed tomography taken after placement of a ventricular catheter with ultrasoundguidance. Although the catheter tip is in adequate position, the burr hole was improperly placed. The patient’s superior sagittal sinus was injured with a subsequent venous infarction. Vascular clips placed to control sinus bleeding are seen.

tip of the catheter was adequately placed within the ventricular system via the use of US, and the shunt has not required revision, poor preoperative planning resulted in a burr hole that was just off the midline, with resultant damage to the superior sagittal sinus and a subsequent venous infarction.

CONCLUSION The results of our study indicate that the use of intraoperative US guidance for the placement of permanent CSF-shunt catheters is associated with a decreased risk of shunt revision at one year. In addition, US guidance was associated with decreased rates of ventricular catheter revision, decreased rates of improperly placed ventricular catheters, and an increased lifespan of properly placed ventricular catheters. Because the intraoperative US is a relatively simple tool to incorporate into one’s surgical repertoire, we advocate for its use in the placement of permanent CSF shunts. REFERENCES 1. Altman NR, Duchowny MS, Jayakar P, Resnick TJ, Alvarez LA, Morrison G: Placement of intracerebral depth electrodes during excisional surgery for epilepsy: value of intraoperative ultrasound. AJNR Am J Neuroradiol 13:254-256, 1992.

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2013.01.039

PEER-REVIEW REPORTS

2. Azeem SS, Origitano TC: Ventricular catheter placement with a frameless neuronavigational system: a 1-year experience. Neurosurgery 60(4 Suppl 2):243-247, 2007; discussion 247-248. 3. Babcock DS, Barr LL, Crone KR: Intraoperative uses of ultrasound in the pediatric neurosurgical patient. Pediatr Neurosurg 18:84-91, 1992. 4. Bierbrauer KS, Storrs BB, McLone DG, Tomita T, Dauser R: A prospective, randomized study of shunt function and infections as a function of shunt placement. Pediatr Neurosurg 16:287-291, 1990. 5. Chandler WF, Knake JE, McGillicuddy JE, Lillehei KO, Silver TM: Intraoperative use of realtime ultrasonography in neurosurgery. J Neurosurg 57:157-163, 1982. 6. Farahmand D, Hilmarsson H, Hogfeldt M, Tisell M: Perioperative risk factors for short-term revision in adult hydrocephalus patients. J Neurol Neurosurg Psychiatry 80:1248-1253, 2009. 7. Hayhurst C, Beems T, Jenkinson MD, Byrne P, Clark S, Kandasamy J, Goodden J, Nandoe Tewarie RD, Mallucci CL: Effect of electromagnetic-navigated shunt placement on failure rates: a prospective multicenter study. J Neurosurg 113: 1273-1278, 2010. 8. Kestle JR, Drake JM, Cochrane DD, Milner R, Walker ML, Abbott R 3rd, Boop FA: Lack of benefit of endoscopic ventriculoperitoneal shunt insertion: a multicenter randomized trial. J Neurosurg 98:284-290, 2003.

US GUIDANCE TO PLACE VENTRICULAR CSF SHUNT CATHETERS

9. Knake JE, Chandler WF, McGillicuddy JE, Silver TM, Gabrielsen TO: Intraoperative sonography for brain tumor localization and ventricular shunt placement. AJR Am J Roentgenol 139: 733-738, 1982.

16. Strowitzki M, Komenda Y, Eymann R, Steudel WI: Accuracy of ultrasound-guided puncture of the ventricular system. Childs Nerv Syst 24:65-69, 2008.

10. Mangano FT, Limbrick DD Jr, Leonard JR, Park TS, Smyth MD: Simultaneous image-guided and endoscopic navigation without rigid cranial fixation: application in infants: technical case report. Neurosurgery 58(4 Suppl 2), 2006; ONSE377; discussion ONS-E377.

17. Strowitzki M, Moringlane JR, Steudel W: Ultrasound-based navigation during intracranial burr hole procedures: experience in a series of 100 cases. Surg Neurol 54:134-144, 2000.

11. Piatt JH Jr, Carlson CV: A search for determinants of cerebrospinal fluid shunt survival: retrospective analysis of a 14-year institutional experience. Pediatr Neurosurg 19:233-241, 1993; discussion 242. 12. Rogers JV 3rd, Shuman WP, Hirsch JH, Lange SC, Howe JF, Burchiel K: Intraoperative neurosonography: application and technique. AJNR Am J Neuroradiol 5:755-760, 1984. 13. Rubin JM, Dohrmann GJ: Use of ultrasonically guided probes and catheters in neurosurgery. Surg Neurol 18:143-148, 1982. 14. Sheth SA, McGirt M, Woodworth G, Wang P, Rigamonti D: Ultrasound guidance for distal insertion of ventriculo-atrial shunt catheters: technical note. Neurol Res 31:280-282, 2009.

18. Whitehead WE, Jea A, Vachhrajani S, Kulkarni AV, Drake JM: Accurate placement of cerebrospinal fluid shunt ventricular catheters with real-time ultrasound guidance in older children without patent fontanelles. J Neurosurg 107:406-410, 2007.

Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 31 January 2012; accepted 10 January 2013; published online 12 January 2013 Citation: World Neurosurg. (2014) 81, 2:397-403. http://dx.doi.org/10.1016/j.wneu.2013.01.039 Journal homepage: www.WORLDNEUROSURGERY.org

15. Shkolnik A, McLone DG: Intraoperative realtime ultrasonic guidance of ventricular shunt placement in infants. Radiology 141:515-517, 1981.

WORLD NEUROSURGERY 81 [2]: 397-403, FEBRUARY 2014

Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2014 Elsevier Inc. All rights reserved.

www.WORLDNEUROSURGERY.org

403

HYDROCEPHALUS

R. WEBSTER CROWLEY ET AL.