January 2005

An 8-month-old boy was admitted to the hospital by his primary physician with intermittent emesis and a bulging anterior fontanel (AF). He had no fever, diarrhea or other new or concerning problems. His past medical history was significant for having a ventriculoperitoneal shunt (VPS) due to holoprosencephaly, and other associated neurological problems. His examination revealed normal vital signs and the bulging AF with widened sutures. He otherwise appeared to be at his baseline. The VPS valve appeared to be slow to refill, but neurosurgical opinion was that it was still functional. A shunt series revealed the shunt to be intact (Figures 1–3), and computed tomography (CT) scan of the head is shown in Figure 4, revealing no interval change since his last CT scan.

The next day, he had a low-grade fever and the shunt was tapped. The cerebrospinal fluid (CSF) analysis revealed no red blood cells and 9,900 white blood cells with 78% segs. The protein was >3 g, and glucose was 1 mg/dL. The Gram’s stain revealed gram-positive cocci. Based on these results, therapy with vancomycin plus ceftriaxone was initiated.

Which of the following would you do next?

1. Replace the VPS with a ventriculostomy
2. Externalize the old shunt
3. Leave the old shunt in place
4. Give antibiotics through the old, externalized VPS

Answer

My answer is A. There are about 10 possible scenarios when one is considering all the shunt options. However, most pediatric infectious disease consultants and pediatric neurosurgeons recommend complete removal of the old device and replacing it with a temporary ventriculostomy (Figure 5) pending clearance of the infection. Alternatively, one could remove the VPS and perform serial ventricular taps every day or every other day to decompress the ventricles until it is safe to place a new device. Ventricular taps are usually done by neurosurgeons but are fairly easy to learn. You identify the lateral angle of the anterior fontanelle as the landmark (Figure 6) and prep the area just as you would for a spinal tap (Figure 7). Using a spinal tap needle with a stylette, you insert the needle at the landmark, directing it toward the medial canthus of the ipsilateral eye (Figure 8). Depending on the depth of the ventricle (degree of hydrocephalus) as determined by the appearance on the CT scan, you can then remove the stylette and collect the fluid for testing and decompression (Figure 9). In the case presented, the needle did not need to be inserted very far. As simple as this procedure is, you should not try it until you have been shown how by a neurosurgeon.

Of course, the antibiotic choices should be adjusted according to the culture results. This patient’s ventricular fluid grew Staphylococcus epidermidis, the most common cause of VPS infections. Because most are methicillin resistant, experts recommend a combination of vancomycin (60 mg/kg/day divided every six hours to start pending levels) and rifampin or gentamicin for synergy against staph. To my knowledge, there are no convincing data for this synergy in humans, as there are in vitro. However, gentamicin may also be reasonable to use to cover for possible gram-negative organisms pending culture results, although we usually use a third-generation cephalosporin for this indication because of its improved blood-brain–barrier penetration.

Once you know the identity of the organism, treatment can be more directed. If vancomycin is being used, you can check levels in the CSF by just taking some out of the reservoir if a drain is in place or sending a sample from the ventricular tap. This helps confirm adequacy of dosing while you follow serum trough levels. Repeat CSF cultures should also be sent every day or two. The duration of therapy until shunt replacement is usually about a week from the first negative culture of the CSF. An additional week or two of therapy is usually needed after shunt replacement depending on the clinical course, causative organism and rate of CSF sterilization.

As more children are surviving the various problems that lead to needing shunts, shunt problems, such as infections, are on the rise.

This can be a very complicated area of pediatrics, and an infectious disease consultation should be considered. Other shunt problems might include allergic or foreign body reactions, break or malfunction, erosion through the skin (Figure 10) or other adjacent structures and abdominal pseudocyst formation. Figure 11 shows the abdominal CT scan of a severely compromised child with a partial intestinal obstruction secondary to a large CSF pseudocyst at the terminal end of the VP shunt.

Regarding the other choices of the management list above, they have all been tried and result in higher complication rates and longer hospitalizations. An excellent review of this topic can be found in The Pediatric Infectious Disease Journal (2002;27[7]:632-636), by Rob Wittler and colleagues. This is the paper I typically use to convince our neurosurgery colleagues to pull the device out when we are dealing with an infected shunt.

The patient presented did well and has remained well with his new shunt.

Acknowledgement and commentary

I would like to thank Dr. Ibrahim Elnihum, pediatric neurosurgeon at Scott & White Hospital, for his assistance with Figures 6, 7, 8 and 9. We are quite fortunate to have an excellent professional relationship with our various surgical colleagues here at Scott & White/Texas A&M University System Health Science Center, College of Medicine. It will make your work easier with a collegial association with your surgeons rather than adversarial, and your patients will benefit as well.

Lastly, please keep our troops in your thoughts. I am concerned that we seem to have entered into a phase of increasing complacency about the war. When it moves off the front page, it’s easier to not think about it. We live about 20 miles from Fort Hood, Texas, the largest U.S. Army installation in the world. Thousands of Fort Hood soldiers are deployed to the Middle East. News of fatalities has become fairly common. In fact, they are found in the obituary section on the day of this writing (page 13 of section B). Obviously, getting out of this with a victory of some sort is of paramount importance, and most military personnel will return alive and in one piece. But we see the effect on the families of these soldiers virtually every day, as we usually have at least one of their children hospitalized in our facility for one reason or another. The vast majority of these parents left behind (mostly mothers), of hospitalized children, do remarkably well, never complaining about anything. But it is easy to read the stress in their personal lives written on their faces. They live in constant fear of the chaplain in dress uniform showing up at their front door.

This time last year, I was asking for names and e-mail addresses of deployed pediatricians, and there was a brisk response. Many of these pediatricians were contacted and responded to our request for information, which was passed on in this column. Their names have changed, but they are still there. Also at this time last year, there was a stunningly different public response to the war. Maybe your town or city is different, but it’s frequently no longer the top news story here, and that’s an unfortunate and dangerous thing, in my opinion. Happy New Year.

For more information:

• James H. Brien, DO, Pediatric Infectious Disease, Scott and White’s Children’s Health Center and Associate Professor of Pediatrics, Texas A&M University, College of Medicine, Temple, Texas. E-mail: jhbrien@aol.com