- Email: [email protected]
Ventriculoperitoneal shunt infection masquerading as an acute surgical abdomen
Ventriculoperitoneal Shunt Infection Masquerading as an Acute Surgical Abdomen Marleta Reynolds, Joseph O. Sherman, and David G. Mclone Chicago, Illinois 9 Shunting of cerebrospinal fluid to the peritoneal cavity has brightened the outlook for children with hydrocephalus. Nine hundred sixty-nine primary ventriculoperitoneal shunts w e r e inserted for hydrocephalus between 1970 and 1981. During this same period, 2 2 0 5 shunt revisions w e r e performed in 847 children, some of whose primary shunt had been inserted prior to 1970 or at other institutions. Nineteen patients with a ventriculoperitoneal shunt infection persented with abdominal pain, fever, and abdominal tenderness; each had acute peritonitis. Three underwent laparotomy with the preoperative diagnosis of appendicitis; however, only infected peritoneal fluid and nonobstructing adhesions w e r e found. A fourth child underwent an unnecessary intestinal resection at another hospital and required prolonged nutritional support and treatment of severe postoperative complications. Fifteen children who presented with an "acute surgical abdomen" w e r e managed with intravenous fluids, gastric decompression, antibiotics, and removal of the intraperitoneal shunt. External ventricular drainage w a s employed until the cerebrospinal fluid was sterile. The shunt was then internalized in the peritoneal cavity. The abdominal signs and symptoms improved after removing the peritoneal tubing in all children. This plan of therapy has eliminated unnecessary laparotomy in those who may require repeated procedures for control of hydrocephalus. INDEX WORDS: Ventriculoperitoneal shunt; hydrocephalus; acute peritonitis.
NFECTION has been reported to develop in
5% to 27% of the ventriculoperitoneal shunts Iinserted for cerebrospinal fluid diversion in patients with hydrocephalus.~-3Prompt diagnosis and treatment are essential because of the morbidity and mortality due to rapidly increasing intracranial pressure and ventriculitis. Occasionally, patients with ventriculoperitoneal shunt infections will present with acute abdominal pain which makes accurate diagnosis more difficult. We are reporting the cases of 19 patients with infected ventriculoperitoneal shunts who presented with fever and abdominal pain and tenderness. The finding of peritonitis prompted a pediatric surgical consultation by the neurosurgeon in each case. Awareness of this unusual presentation of shunt infection facilitates appropriate treatment and obviates unnecessary laparotomy. Journal of Pediatric Surgery, Vol. 18, No. 6 (December), 1983
MATERIALS AND METHODS Nine hundred sixty-nine primary ventriculoperitoneal shunts were inserted for hydrocephalus between 1970 and 1981 at The Children's Memorial Hospital in Chicago, Illinois. In addition, 2,205 shunt revisions were performed in 847 children for shunt malfunction or infection. Nineteen patients with ventriculoperitoneal shunt infection presented with signs and symptoms of an acute surgical abdomen, and four underwent exploratory laparotomy. There were equal numbers of boys and girls aged 8 months to 13 years. Congenital anomalies were responsible for hydrocephalus in a majority of the patients. An average of four years had elpased between the last shunt revision and the present admission for shunt infection. Most of our patients had had abdominal symptoms for several days prior to admission. The most frequent physical findings were fever and abdominal tenderness. All had peritoneal signs and 58% had localized tenderness (Table 1). On the basis of only the abdominal examination, the diagnosis of an acute surgical abdomen necessitating laparotomy would have been made in all cases, Fifty percent of the children exhibited a change in neurologic status or signs of increased intraeranial pressure, such as lethargy and irritability. Leukocytosis with a left shift was present in 74% of the patients. Abdominal x-rays were reported as normal in 69%, suggestive of mechanical obstruction in 12%, and suggestive of ileus in 19% (Table 1). The shunt tip was localized in the right lower quadrant in 44%. There was no correlation between the location of the tip of the shunt tubing and the location of the abdominal tenderness. Fifteen of the patients were treated with external ventrieular drainage, antibiotics, and intravenous fluids. The remaining four underwent an exploratory laparotomy followed by external ventricular drainage. The average time between admission and operation for all patients was 42 hours. Culture and sensitivity studies of the shunt tubing or cerebrospinal fluid were available in 90% of the patients. Grampositive organisms were isolated in 65%; gram-negative organisms, in 23%. Symptoms resolved within 31 hours after externalization of the shunt tubing. Purulent fluid and adhesions were found intraoperatively in each patient who underwent exploratory
From the Divisions of Pediatric Surgery and Neurological Surgery, The Children's Memorial Hospital, The Department of Surgery, The Northwestern University Medical School, Chicago, IL. Presented before the 14th Annual Meeting of the American Pediatric Surgical Association, Hilton Head Island, South Carolina, May 4.-7, 1983. Address reprint request to Joseph O. Sherman, MD, The Children's Memorial Hospital, 2300 Children's Plaza, Chicago, IL 60614. 9 1983 by Grune & Straiten, Inc. 00 22/3468/8 3/1806~06 2501.0(9/0 951
952
REYNOLDS, SHERMAN, AND MCLONE
Table 1. Clinical Assessment of 19 Patients W i t h Ventriculoperitoneal Shunts and Acute Abdomen
Presenting Symptoms Fever Abdominal pain Nausea/vomiting Anorexia Bloating Diarrhea Physical Findings Peritonitis Localized tenderness Generalized tenderness Abdominal distension Altered neurologic exam White Blood Cell Count Leukocytosis with left shift Normal Abdominal Roentgenograms Normal Ileus Mechanical obstruction
14 14 9 5 5 3 19 11 8 8 9
14 5 11 3 2
laparotomy, but no source of sepsis could be identified. The number of hospital days per admission ranged from 5 to 47 days, with a mean of 32 days. One patient died of ventriculitis and pneumonia. C A S E REPORTS
MR, 3-year-boy with postmeningitis hydrocephalus had undergone shunt revision 2 months prior to admission. He presented with fever and abdominal pain and tenderness. His white blood cell count was 12,600 with 72% polys. Abdominal x-ray films were normal. Acute appendicitis was suspected, and an exploratory laparotomy was performed. Purulent fluid and adhesions were identified, but the appendix was normal and no other intraabdominal pathologic changes were found. His ventriculoperitoneal shunt was externalized, and antibiotic therapy was started. He improved within 24 hours. MG, 13-year-old boy with congenital hydrocephalus, had last undergone shunt revision 3 months before admission. He presented with a one-day history of fever, abdominal pain, and tenderness in the right lower quadrant. His white blood cell count was 16,200 with 80% polys and 11% bands. The abdominal x-ray films were normal. His shunt was externalized, and antibiotics were administered. His symptoms resolved within 12 hours. EL, a 29-month-old boy, had a Dandy-Walker cyst drained 2 years prior to admission. Eight days prior to admission, fever, increasing irritability and slight right lower quadrant tenderness developed. Three days after admission, an intravenous pyelogram revealed a dilated right collecting system. A mass in the right lower quadrant was thought to represent a cerebrospinal fluid cyst. External ventricular drainage was performed to relieve increasing intracranial pressure. Cerebrospinal fluid culture was positive for grampositive cocci. His neurologic condition continued to deteriorate; and he was flaccid, and his pupils became dilated. Cardiac arrest followed. Blood cultures were negative and
autopsy revealed bronchopneumonia and a well-localized right lower quadrant abscess. No evidence of overwhelming sepsis was identified. JS, a 7-year-old girl with myelomeningocele, presented with nausea and abdominal pain and tenderness 2 years after her last ventriculoperitoneal shunt revision. Purulent fluid and adhesions were found at exploratory laparotomy at another institution. The pathologic condition was unclear to the surgeon, and 30% to 50% of the small bowel and ascending colon was resected. Her postoperative course was complicated by an enterocutaneous fistula, subphrenic abscess, and short-gut syndrome. She was transferred to The Children's Memorial Hospital for treatment of these complications. She was readmitted because of abdominal pain 1 year after the initial operation and again 2 years later; infected intraabdominal cysts were drained on both occasions. These case reports demonstrate the appropriate management of these patients and the devastating complications that can follow inappropriate treatment and ill-advised abdominal surgery. DISCUSSION
I n f e c t i o n of c e r e b r o s p i n a l fluid s h u n t s is m o r e c o m m o n in t h e very y o u n g , especially w h e n perf o r m e d for c o n g e n i t a l h y d r o c e p h a l u s . T h e 13% m o r t a l i t y r a t e a s s o c i a t e d with v e n t r i c u l o p e r i t o neal s h u n t i n f e c t i o n has p r o m p t e d several studies to i d e n t i f y risk factors. ~ O n c e believed to be s e c o n d a r y to t r a n s i e n t b a c t e r e m i a , c o n t a m i n a tion f r o m the skin in t h e p e r i o p e r a t i v e period is the m o r e likely cause. 2 I n d i v i d u a l s u r g i c a l techn i q u e has b e e n i m p l i c a t e d as a risk factor a n d Venes has d e s c r i b e d a m e t i c u l o u s o p e r a t i v e techn i q u e to i n s u r e sterility w h e n i n s e r t i n g v e n t r i c u l o p e r i t o n e a l s h u n t s . 3 N e i t h e r the m a t e r i a l s used for the s h u n t t u b i n g n o r the b r a n d of s h u n t t u b i n g were f o u n d to i n f l u e n c e the s h u n t infection rate. W e have b e e n u s i n g the R a i m o n d i s h u n t a l m o s t exclusively for m o r e t h a n 7 years. Staphylococcus epidermidis have b e e n t h e m o s t common infecting organisms and prophylactic a n t i b o d i e s have r e s u l t e d in equivocal success in p r e v e n t i n g infection. W h e n t h e s h u n t i n f e c t i o n m a n i f e s t s as b a c t e r i a l peritonitis, it is t h o u g h t to result f r o m c o n t a m i n a t i o n of the p e r i t o n e a l cavity with i n f e c t e d c e r e b r o s p i n a l fluid. T h e c o n f u sion s u r r o u n d i n g the e x a c t source of the p e r i t o n i tis has p r o m p t e d this r e p o r t of 19 c h i l d r e n w i t h ventriculoperitoneal shunt infection and an acute surgical a b d o m e n . T h e s e p a t i e n t s p r e s e n t e d as p e r p l e x i n g d i a g nostic p r o b l e m s . E a c h d e m o n s t r a t e d t h e signs a n d s y m p t o m s of a c u t e a p p e n d i c i t i s . A l l w o u l d have u n d e r g o n e u n n e c e s s a r y l a p a r o t o m y if t h e
953
VP SHUNT INFECTIONS
surgeon had not been aware that ventriculoperitoneal shunt infections can mimic an acute surgical abdomen. In contrast, most shunt infections are manifested classically by nonspecific complaints including fever, nausea, vomiting, and malaise. In addition, shunt malfunctions may result in signs of increased intracranial pressure and altered neurologic status. The recommended treatment includes externalization of the shunt with replacement after intensive antibiotic theraPy. Does the surgeon proceed with an exploratory laparotomy in a patient with classic signs or symptoms of an acute surgical abdomen or can the process be resolved with externalization of the ventriculoperitoneal shunt? From the information we have collected it appears that a correct diagnosis cannot be made with confidence given the history and physical examination, x-ray, and laboratory studies currently available. Intraabdominal complications of ventriculoperitoneal shunts have been reported to occur in up to 24% of infants and children who undergo ventriculoperitoneal shunt placement for hydrocephalus. Grosfeld et al reported complications that included inguinal hernias, perforated viscus, and peritoneal cyst formation. 4 Rekate et al described two distinct groups of patients with intraabdominal complications of ventriculoperitoneal shunts. 5 The first group was similar to that of Grosfeid et al in that it included two patients
with perforated viscus; the remaining six presented with peritonitis and were found to have infected ventriculoperitoneal shunts rather than a primary intraabdominal pathologic condition. Others have reported individual cases of bacterial peritonitis resulting from infected ventriculoperitoneal shunts. 6-~ In our series of patients, the correct diagnosis was difficult to determine from the data available on each patient, and unnecessary laparotomies were performed. Ventriculoperitoneal shunting has become the procedure of choice for CSF diversion in infants and children. We have reported the cases of 19 children with infected ventriculoperitoneal shunts masquerading as an acute surgical abdomen. The correct diagnosis was unclear, and exploratory laparotomies were performed in four patients. Given the obvious similarities in presentation and the absence of differentiating criteria, we recommend immediate externalization of the ventriculo-peritoneal shunt. Antibiotic therapy to combat staphylococcal organisms should be instituted in combination with an aminoglycoside pending results of culture and sensitivity studies of the cerebrospinal fluid and shunt tip. If the abdominal findings have not improved in 6 to 8 hours, the patient should be reevaluated and an exploratory laparotomy should be considered. We have not performed any unnecessary laparotomies since this plan of treatment was adopted.
REFERENCES 1. George R, Leibrock L, Epstein M: Long-term analysis of cerebrospinal fluid shunt infections. J Neurosurg 51:804811, 1979 2. Schoenbaum CS, Gardner P, Shillito J: Infections of cerebrospinal fluid shunts: Epidemiology, clinical manifestations and therapy. J Infect Dis 131:543-552, 1975 3. Venes JL: Control of shunt infections. J Neurosurg 45:311-314, 1976 4. Grosfeld JL, Cooney DR, Smith J e t al: Intra-abdominal complications following ventriculo-peritoneal shunt procedures. Pediatrics 54:791-796, 1974
5. Rekate HL, Yonas H, While RJ et al: The acute abdomen in patients with ventriculo-peritoneal shunts. Surg Neurol 11:442-445, 1979 6. Hubschmann OR, Countee RW: Gram-positive peritonitis in patients with infected ventriculo-peritoneal shunts. Surg Gynecol Obstet 149:69-71, 1979 7. Leibrock L, Baker R, Vematsu S: Simulated acute appendicitis secondary to ventriculo-peritoneal shunts. Surg Neurol 4:481-482, 1975 8. Tchirkow G, Verhagen AD: Bacterial peritonitis in patients with ventriculo-peritoneal shunts. J Pediatr Surg 14:182-184, 1979
Discussion Bradley Rogers (Charlottesville, Va.): This is a problem which we are seeing more frequently. Several years ago we presented our use of laparoscopy in the treatment of the peritoneal cysts in
these patients. One of the things that we found in that relatively small group of patients was that if you reinsert the shunt after an infection in the peritoneum, if gets infected again and ceases to
954
function. I was interested that the authors recommend the reinsertion of a ventriculoperitoneal shunt under these circumstances and I wonder if they have longterm follow-up on those patients as to whether they have gotten secondary infection or secondary obstruction in that setting? The only disagreement I think that I would have with the authors recommendations deals with their recommendation for immediate externalization of these shunts. A functioning VP shunt is gold for these children and to immediately externalize it on the presumption of CSF infection or peritonitis from infected CSF may be excessively aggressive. We've preferred to pursue a course of tapping the shunt doing gram stains or cultures. The gram stains probably are more important than the culture because occasionally the culture will be negative initially. So far we have not been misled by the gram stain and I'd be interested in the authors' experience in this regard. The question I have is whether the author's recommend prophylactic use of antibiotics in patients with VP shunts, much as we do in patients with cardiac valves or valve diseases. Although certainly most of these infections are introduced at the time of surgery, many of them arise considerably later, and I wonder if we can prevent some of these infections with more vigorous use of prophyllactus. Neil Feins (Brighton, MA): I would like to ask a question. With pressure from the neurosurgeon, the pediatrician, and the surgical resident feeling an acute abdomen, in order to avoid an unnecessary laparotomy, have you used ultrasound or a barium enema for a diagnosis prior to making any surgical maneuvers or suggesting removal of the shunt? Bruce McGovern (San Antonio): I have had trouble convincing our neurosurgeons that the shunt is infected. They will often tap the shunt reservoir. When the gram stain and culture of the fluid is negative they infer that the cause must be appendicitis. I have found it helpful to perform a needle paracentesis of the abdomen. If the
REYNOLDS, SHERMAN, AND MCLONE
abdominal fluid has gram positive cocci, shunt infection is likely. Thomas Lobe (Galveston): I'd like to ask the authors their advice with regards to abdominal masses in these children with acute abdomens and V-P shunts. 1 have had one patient recently who had an abdominal mass. His cutures grew out Candida albincans. We treated him for 8 weeks with amphotericin B and his mass went away without requiring an operation. I had a more confusing case when I was a resident in Columbus. We had a girl with a primary Pinealoma who later presented with a large abdominal mass and an acute abdomen. I assumed that it was infection, but it was discovered to be a yolk sac tumor which may have seaded from the Pinealoma. I think children with V-P shunts who present with an abnormal mass and an acute abdomen are a different group of patients and I'd like to ask the authors for their advice. Marleta Reynolds (closing): None of the patients with shunt-related peritonitis treated with externalization and systemic antibiotics have returned with abdominal complaints or complications. The use of parenteral antibiotics alone to treat these patients with peritonitis has been abandoned by most neurosurgeons because immediate externalization of the shunt has resulted in a marked decrease in morbidity and mortality. The results of the use of prophylactic antibiotics in several clinical studies have been equivocal. Ultrasound has identified cerebrospinal fluid collections, abscesses, and other abdominal masses in some of these children. We have not seen the clinical indication for barium enema or peritoneal tap. Even when gram-negative organisms were identified in the cerebrospinal fluid, rapid clinical improvement made it obvious that there was no primary intraabdominal pathology. Shunts are also externalized in patients with intraabdominal cerebrospinal fluid collections, but the patients usually do not present with an acute surgical abdomen.
NFECTION has been reported to develop in
5% to 27% of the ventriculoperitoneal shunts Iinserted for cerebrospinal fluid diversion in patients with hydrocephalus.~-3Prompt diagnosis and treatment are essential because of the morbidity and mortality due to rapidly increasing intracranial pressure and ventriculitis. Occasionally, patients with ventriculoperitoneal shunt infections will present with acute abdominal pain which makes accurate diagnosis more difficult. We are reporting the cases of 19 patients with infected ventriculoperitoneal shunts who presented with fever and abdominal pain and tenderness. The finding of peritonitis prompted a pediatric surgical consultation by the neurosurgeon in each case. Awareness of this unusual presentation of shunt infection facilitates appropriate treatment and obviates unnecessary laparotomy. Journal of Pediatric Surgery, Vol. 18, No. 6 (December), 1983
MATERIALS AND METHODS Nine hundred sixty-nine primary ventriculoperitoneal shunts were inserted for hydrocephalus between 1970 and 1981 at The Children's Memorial Hospital in Chicago, Illinois. In addition, 2,205 shunt revisions were performed in 847 children for shunt malfunction or infection. Nineteen patients with ventriculoperitoneal shunt infection presented with signs and symptoms of an acute surgical abdomen, and four underwent exploratory laparotomy. There were equal numbers of boys and girls aged 8 months to 13 years. Congenital anomalies were responsible for hydrocephalus in a majority of the patients. An average of four years had elpased between the last shunt revision and the present admission for shunt infection. Most of our patients had had abdominal symptoms for several days prior to admission. The most frequent physical findings were fever and abdominal tenderness. All had peritoneal signs and 58% had localized tenderness (Table 1). On the basis of only the abdominal examination, the diagnosis of an acute surgical abdomen necessitating laparotomy would have been made in all cases, Fifty percent of the children exhibited a change in neurologic status or signs of increased intraeranial pressure, such as lethargy and irritability. Leukocytosis with a left shift was present in 74% of the patients. Abdominal x-rays were reported as normal in 69%, suggestive of mechanical obstruction in 12%, and suggestive of ileus in 19% (Table 1). The shunt tip was localized in the right lower quadrant in 44%. There was no correlation between the location of the tip of the shunt tubing and the location of the abdominal tenderness. Fifteen of the patients were treated with external ventrieular drainage, antibiotics, and intravenous fluids. The remaining four underwent an exploratory laparotomy followed by external ventricular drainage. The average time between admission and operation for all patients was 42 hours. Culture and sensitivity studies of the shunt tubing or cerebrospinal fluid were available in 90% of the patients. Grampositive organisms were isolated in 65%; gram-negative organisms, in 23%. Symptoms resolved within 31 hours after externalization of the shunt tubing. Purulent fluid and adhesions were found intraoperatively in each patient who underwent exploratory
From the Divisions of Pediatric Surgery and Neurological Surgery, The Children's Memorial Hospital, The Department of Surgery, The Northwestern University Medical School, Chicago, IL. Presented before the 14th Annual Meeting of the American Pediatric Surgical Association, Hilton Head Island, South Carolina, May 4.-7, 1983. Address reprint request to Joseph O. Sherman, MD, The Children's Memorial Hospital, 2300 Children's Plaza, Chicago, IL 60614. 9 1983 by Grune & Straiten, Inc. 00 22/3468/8 3/1806~06 2501.0(9/0 951
952
REYNOLDS, SHERMAN, AND MCLONE
Table 1. Clinical Assessment of 19 Patients W i t h Ventriculoperitoneal Shunts and Acute Abdomen
Presenting Symptoms Fever Abdominal pain Nausea/vomiting Anorexia Bloating Diarrhea Physical Findings Peritonitis Localized tenderness Generalized tenderness Abdominal distension Altered neurologic exam White Blood Cell Count Leukocytosis with left shift Normal Abdominal Roentgenograms Normal Ileus Mechanical obstruction
14 14 9 5 5 3 19 11 8 8 9
14 5 11 3 2
laparotomy, but no source of sepsis could be identified. The number of hospital days per admission ranged from 5 to 47 days, with a mean of 32 days. One patient died of ventriculitis and pneumonia. C A S E REPORTS
MR, 3-year-boy with postmeningitis hydrocephalus had undergone shunt revision 2 months prior to admission. He presented with fever and abdominal pain and tenderness. His white blood cell count was 12,600 with 72% polys. Abdominal x-ray films were normal. Acute appendicitis was suspected, and an exploratory laparotomy was performed. Purulent fluid and adhesions were identified, but the appendix was normal and no other intraabdominal pathologic changes were found. His ventriculoperitoneal shunt was externalized, and antibiotic therapy was started. He improved within 24 hours. MG, 13-year-old boy with congenital hydrocephalus, had last undergone shunt revision 3 months before admission. He presented with a one-day history of fever, abdominal pain, and tenderness in the right lower quadrant. His white blood cell count was 16,200 with 80% polys and 11% bands. The abdominal x-ray films were normal. His shunt was externalized, and antibiotics were administered. His symptoms resolved within 12 hours. EL, a 29-month-old boy, had a Dandy-Walker cyst drained 2 years prior to admission. Eight days prior to admission, fever, increasing irritability and slight right lower quadrant tenderness developed. Three days after admission, an intravenous pyelogram revealed a dilated right collecting system. A mass in the right lower quadrant was thought to represent a cerebrospinal fluid cyst. External ventricular drainage was performed to relieve increasing intracranial pressure. Cerebrospinal fluid culture was positive for grampositive cocci. His neurologic condition continued to deteriorate; and he was flaccid, and his pupils became dilated. Cardiac arrest followed. Blood cultures were negative and
autopsy revealed bronchopneumonia and a well-localized right lower quadrant abscess. No evidence of overwhelming sepsis was identified. JS, a 7-year-old girl with myelomeningocele, presented with nausea and abdominal pain and tenderness 2 years after her last ventriculoperitoneal shunt revision. Purulent fluid and adhesions were found at exploratory laparotomy at another institution. The pathologic condition was unclear to the surgeon, and 30% to 50% of the small bowel and ascending colon was resected. Her postoperative course was complicated by an enterocutaneous fistula, subphrenic abscess, and short-gut syndrome. She was transferred to The Children's Memorial Hospital for treatment of these complications. She was readmitted because of abdominal pain 1 year after the initial operation and again 2 years later; infected intraabdominal cysts were drained on both occasions. These case reports demonstrate the appropriate management of these patients and the devastating complications that can follow inappropriate treatment and ill-advised abdominal surgery. DISCUSSION
I n f e c t i o n of c e r e b r o s p i n a l fluid s h u n t s is m o r e c o m m o n in t h e very y o u n g , especially w h e n perf o r m e d for c o n g e n i t a l h y d r o c e p h a l u s . T h e 13% m o r t a l i t y r a t e a s s o c i a t e d with v e n t r i c u l o p e r i t o neal s h u n t i n f e c t i o n has p r o m p t e d several studies to i d e n t i f y risk factors. ~ O n c e believed to be s e c o n d a r y to t r a n s i e n t b a c t e r e m i a , c o n t a m i n a tion f r o m the skin in t h e p e r i o p e r a t i v e period is the m o r e likely cause. 2 I n d i v i d u a l s u r g i c a l techn i q u e has b e e n i m p l i c a t e d as a risk factor a n d Venes has d e s c r i b e d a m e t i c u l o u s o p e r a t i v e techn i q u e to i n s u r e sterility w h e n i n s e r t i n g v e n t r i c u l o p e r i t o n e a l s h u n t s . 3 N e i t h e r the m a t e r i a l s used for the s h u n t t u b i n g n o r the b r a n d of s h u n t t u b i n g were f o u n d to i n f l u e n c e the s h u n t infection rate. W e have b e e n u s i n g the R a i m o n d i s h u n t a l m o s t exclusively for m o r e t h a n 7 years. Staphylococcus epidermidis have b e e n t h e m o s t common infecting organisms and prophylactic a n t i b o d i e s have r e s u l t e d in equivocal success in p r e v e n t i n g infection. W h e n t h e s h u n t i n f e c t i o n m a n i f e s t s as b a c t e r i a l peritonitis, it is t h o u g h t to result f r o m c o n t a m i n a t i o n of the p e r i t o n e a l cavity with i n f e c t e d c e r e b r o s p i n a l fluid. T h e c o n f u sion s u r r o u n d i n g the e x a c t source of the p e r i t o n i tis has p r o m p t e d this r e p o r t of 19 c h i l d r e n w i t h ventriculoperitoneal shunt infection and an acute surgical a b d o m e n . T h e s e p a t i e n t s p r e s e n t e d as p e r p l e x i n g d i a g nostic p r o b l e m s . E a c h d e m o n s t r a t e d t h e signs a n d s y m p t o m s of a c u t e a p p e n d i c i t i s . A l l w o u l d have u n d e r g o n e u n n e c e s s a r y l a p a r o t o m y if t h e
953
VP SHUNT INFECTIONS
surgeon had not been aware that ventriculoperitoneal shunt infections can mimic an acute surgical abdomen. In contrast, most shunt infections are manifested classically by nonspecific complaints including fever, nausea, vomiting, and malaise. In addition, shunt malfunctions may result in signs of increased intracranial pressure and altered neurologic status. The recommended treatment includes externalization of the shunt with replacement after intensive antibiotic theraPy. Does the surgeon proceed with an exploratory laparotomy in a patient with classic signs or symptoms of an acute surgical abdomen or can the process be resolved with externalization of the ventriculoperitoneal shunt? From the information we have collected it appears that a correct diagnosis cannot be made with confidence given the history and physical examination, x-ray, and laboratory studies currently available. Intraabdominal complications of ventriculoperitoneal shunts have been reported to occur in up to 24% of infants and children who undergo ventriculoperitoneal shunt placement for hydrocephalus. Grosfeld et al reported complications that included inguinal hernias, perforated viscus, and peritoneal cyst formation. 4 Rekate et al described two distinct groups of patients with intraabdominal complications of ventriculoperitoneal shunts. 5 The first group was similar to that of Grosfeid et al in that it included two patients
with perforated viscus; the remaining six presented with peritonitis and were found to have infected ventriculoperitoneal shunts rather than a primary intraabdominal pathologic condition. Others have reported individual cases of bacterial peritonitis resulting from infected ventriculoperitoneal shunts. 6-~ In our series of patients, the correct diagnosis was difficult to determine from the data available on each patient, and unnecessary laparotomies were performed. Ventriculoperitoneal shunting has become the procedure of choice for CSF diversion in infants and children. We have reported the cases of 19 children with infected ventriculoperitoneal shunts masquerading as an acute surgical abdomen. The correct diagnosis was unclear, and exploratory laparotomies were performed in four patients. Given the obvious similarities in presentation and the absence of differentiating criteria, we recommend immediate externalization of the ventriculo-peritoneal shunt. Antibiotic therapy to combat staphylococcal organisms should be instituted in combination with an aminoglycoside pending results of culture and sensitivity studies of the cerebrospinal fluid and shunt tip. If the abdominal findings have not improved in 6 to 8 hours, the patient should be reevaluated and an exploratory laparotomy should be considered. We have not performed any unnecessary laparotomies since this plan of treatment was adopted.
REFERENCES 1. George R, Leibrock L, Epstein M: Long-term analysis of cerebrospinal fluid shunt infections. J Neurosurg 51:804811, 1979 2. Schoenbaum CS, Gardner P, Shillito J: Infections of cerebrospinal fluid shunts: Epidemiology, clinical manifestations and therapy. J Infect Dis 131:543-552, 1975 3. Venes JL: Control of shunt infections. J Neurosurg 45:311-314, 1976 4. Grosfeld JL, Cooney DR, Smith J e t al: Intra-abdominal complications following ventriculo-peritoneal shunt procedures. Pediatrics 54:791-796, 1974
5. Rekate HL, Yonas H, While RJ et al: The acute abdomen in patients with ventriculo-peritoneal shunts. Surg Neurol 11:442-445, 1979 6. Hubschmann OR, Countee RW: Gram-positive peritonitis in patients with infected ventriculo-peritoneal shunts. Surg Gynecol Obstet 149:69-71, 1979 7. Leibrock L, Baker R, Vematsu S: Simulated acute appendicitis secondary to ventriculo-peritoneal shunts. Surg Neurol 4:481-482, 1975 8. Tchirkow G, Verhagen AD: Bacterial peritonitis in patients with ventriculo-peritoneal shunts. J Pediatr Surg 14:182-184, 1979
Discussion Bradley Rogers (Charlottesville, Va.): This is a problem which we are seeing more frequently. Several years ago we presented our use of laparoscopy in the treatment of the peritoneal cysts in
these patients. One of the things that we found in that relatively small group of patients was that if you reinsert the shunt after an infection in the peritoneum, if gets infected again and ceases to
954
function. I was interested that the authors recommend the reinsertion of a ventriculoperitoneal shunt under these circumstances and I wonder if they have longterm follow-up on those patients as to whether they have gotten secondary infection or secondary obstruction in that setting? The only disagreement I think that I would have with the authors recommendations deals with their recommendation for immediate externalization of these shunts. A functioning VP shunt is gold for these children and to immediately externalize it on the presumption of CSF infection or peritonitis from infected CSF may be excessively aggressive. We've preferred to pursue a course of tapping the shunt doing gram stains or cultures. The gram stains probably are more important than the culture because occasionally the culture will be negative initially. So far we have not been misled by the gram stain and I'd be interested in the authors' experience in this regard. The question I have is whether the author's recommend prophylactic use of antibiotics in patients with VP shunts, much as we do in patients with cardiac valves or valve diseases. Although certainly most of these infections are introduced at the time of surgery, many of them arise considerably later, and I wonder if we can prevent some of these infections with more vigorous use of prophyllactus. Neil Feins (Brighton, MA): I would like to ask a question. With pressure from the neurosurgeon, the pediatrician, and the surgical resident feeling an acute abdomen, in order to avoid an unnecessary laparotomy, have you used ultrasound or a barium enema for a diagnosis prior to making any surgical maneuvers or suggesting removal of the shunt? Bruce McGovern (San Antonio): I have had trouble convincing our neurosurgeons that the shunt is infected. They will often tap the shunt reservoir. When the gram stain and culture of the fluid is negative they infer that the cause must be appendicitis. I have found it helpful to perform a needle paracentesis of the abdomen. If the
REYNOLDS, SHERMAN, AND MCLONE
abdominal fluid has gram positive cocci, shunt infection is likely. Thomas Lobe (Galveston): I'd like to ask the authors their advice with regards to abdominal masses in these children with acute abdomens and V-P shunts. 1 have had one patient recently who had an abdominal mass. His cutures grew out Candida albincans. We treated him for 8 weeks with amphotericin B and his mass went away without requiring an operation. I had a more confusing case when I was a resident in Columbus. We had a girl with a primary Pinealoma who later presented with a large abdominal mass and an acute abdomen. I assumed that it was infection, but it was discovered to be a yolk sac tumor which may have seaded from the Pinealoma. I think children with V-P shunts who present with an abnormal mass and an acute abdomen are a different group of patients and I'd like to ask the authors for their advice. Marleta Reynolds (closing): None of the patients with shunt-related peritonitis treated with externalization and systemic antibiotics have returned with abdominal complaints or complications. The use of parenteral antibiotics alone to treat these patients with peritonitis has been abandoned by most neurosurgeons because immediate externalization of the shunt has resulted in a marked decrease in morbidity and mortality. The results of the use of prophylactic antibiotics in several clinical studies have been equivocal. Ultrasound has identified cerebrospinal fluid collections, abscesses, and other abdominal masses in some of these children. We have not seen the clinical indication for barium enema or peritoneal tap. Even when gram-negative organisms were identified in the cerebrospinal fluid, rapid clinical improvement made it obvious that there was no primary intraabdominal pathology. Shunts are also externalized in patients with intraabdominal cerebrospinal fluid collections, but the patients usually do not present with an acute surgical abdomen.