Ventriculoperitoneal (VP) shunt malfunctions and complications are common presentations to pediatric emergency departments (ED), as is abdominal pain in adolescent females; however, abdominal pain is not usually a symptom of shunt malfunction. The usual presenting features of VP shunt malfunctions are headache and vomiting, although the literature cites multiple unusual presentations, including ascites, abdominal pseudocysts, scrotal edema, and nonvisceral penetrations. We report a case of a 16-year-old female patient with a VP shunt who presented to the pediatric ED with headache, abdominal pain, and profuse clear vaginal discharge. At laparoscopy, the distal end of the patient’s VP shunt was found to be draining into a loculated mass encasing a right hydrosalpinx.
A.M., a 16-year-old white female, presented to the Medical University of South Carolina pediatric ED in late July 1999 with a 2-week history of increasingly profuse clear vaginal discharge and a 1-day history of severe lower abdominal pain, headache, nausea, and fever to 101.2°F. She denied vomiting, diarrhea, constipation, or urinary symptoms. Her last menstrual period was 3 weeks prior to presentation, and the patient denied any sexual activity. About 1 week prior to the current admission, the patient was seen by a gynecologist for the vaginal discharge. She had a normal pelvic examination, normal vaginal secretion microscopy, and negative endocervical cultures. The remainder of her review of systems was negative.
Relevant past medical history included VP shunt placement in January 1999 for pseudotumor cerebri with persistent occipital headaches. The surgery followed failure of conservative management over several months, including treatment with lumbar puncture, and medications including diuretics. The patient underwent two shunt revisions for protruding scalp sutures. At a different institution, she had also had a diagnostic laparoscopy with incidental appendectomy in 1996, and a subsequent laparoscopic right ovarian cystectomy in March 1999 at our institution. Oral contraceptives were prescribed for suppression of ovarian cyst formation at that time, but were discontinued a short time afterward by the patient herself.
On examination, the patient was a well-developed teenager, alert and cooperative, but in moderate discomfort from abdominal pain and the marked amount of vaginal discharge. Vital signs were normal. Her abdominal examination revealed mild, diffuse tenderness with voluntary guarding but without rebound. The abdomen was soft and nondistended with normal bowel sounds. There was no hepatosplenomegaly or other palpable mass. Most remarkable was the large pool of clear vaginal discharge that soaked the examination table. Results of the rest of the initial examination, including head, shunt reservoir, pupil, and neurologic examinations, were normal.
Pelvic examination revealed normal external genitalia, normal vaginal tissues, a small nulliparous cervix, and profuse, clear, watery cervical discharge that formed a pool in the posterior vaginal fornix. Bimanual examination revealed moderate cervical motion tenderness. Palpation of the pelvic structures was limited by voluntary guarding; however, a fullness was noted in the right adnexa. Gram stain and wet mount microscopy revealed normal epithelial cells and no evidence of infection. Potassium hydroxide preparation (KOH prep) was negative for fungal elements.
A transvaginal ultrasound revealed a normal-sized uterus; however, the endometrial cavity contained sonolucent fluid throughout the fundus, body, cervix, and both cornu. The amount of fluid distension fluctuated during the study. Free fluid was also noted in the peritoneal cavity. The right ovary measured 4.5 × 2.5 × 3.5 cm and contained multiple cysts, at least one of which contained septations and internal debris. This was believed to represent a hemorrhagic cyst. The left ovary was 2.7 × 1.2 × 2.2 cm with a normal appearance. Both ovaries were well perfused.
Laboratory testing revealed a normal shunt series, with distal tip overlying the right true pelvis, negative serum pregnancy test, normal basic metabolic panel, normal peripheral white blood cell count, sedimentation rate of 6l mm/h, and albumin 3.4 mg/dL. Urinalysis showed trace protein, trace blood, moderate mucus, and moderate calcium oxalate crystals. Shunt fluid revealed no white or red blood cells, glucose 70 mg/dL, and protein 10 mg/dL. The vaginal fluid showed a creatinine of 0.8 mg/dL, glucose of 48 mg/dL, and no protein.
The patient was admitted overnight for observation under the diagnosis of hemorrhagic right ovarian cyst. Her pain and amount of vaginal discharge were somewhat improved in the morning, and she was discharged on narcotic and nonsteroidal pain medications with follow-up in the gynecology clinic 1 week later. At the time of follow-up, the patient reported continued profuse vaginal discharge and increased pain, and she was admitted for surgical management.
Diagnostic laparoscopy revealed a normal uterus, left fallopian tube, and left ovary. There was a large right hydrosalpinx without identifiable fimbria, which was also involved with dense omental adhesions (Fig. 1). The shunt tip was not initially visualized. Upon lysis of the omental adhesions, a rush of clear fluid was released from the loculated mass of omentum and right adnexal structures. The distal tip of the VP shunt was noted to be within this cystic loculation, near the distal fallopian tube. The right ovary was then visualized and noted to be entirely normal. A right salpingectomy was performed, and the tip of the VP shunt was freed. Postoperatively, the patient received 48 hours of intravenous antibiotics, her pain subsided, and her discharge resolved immediately. Over 1 year out, she remains asymptomatic and has had no other shunt or gynecologic problems. Due to the right salpingectomy and remaining right ovary, the gynecology department continues to follow her closely for fertility issues and the increased risk of ectopic pregnancy.
Abdominal pain in teenage female patients usually implies a large differential but does not typically include VP shunt malfunction. Likewise, VP shunt malfunction can present in a myriad of ways, but it does not commonly present with vaginal discharge. Both are frequent evaluations in a pediatric ED.
Important considerations in the evaluation of any type of vaginal discharge associated with abdominal pain in an adolescent patient include infectious etiologies, particularly sexually transmitted infections, and pelvic inflammatory disease. Profuse, watery discharge is not a common presentation, but it is associated with several serious conditions, including fallopian tube carcinoma, cervical carcinoma, ectopic ureter, or fistulous tracts between the urinary and reproductive tracts (1,2). In this case, urinary tract fistula was ruled out by the low creatinine of the vaginal fluid. Although an abnormal connection between the distal tip of the VP shunt and the genitourinary structures was a concern in this patient, there is no simple clinical test to discriminate cerebrospinal fluid (CSF) mixed with peritoneal fluid and cervicovaginal secretions.
A comparison of diagnostic modalities therefore deserves attention. In this case, the patient had risk factors for adhesions (the two previous laparoscopic procedures, most notably the right ovarian cystectomy and appendectomy, and shunt placement itself) that led to suspicion of an abnormal abdominopelvic process. It is of interest in this case that even a transvaginal ultrasound in combination with a shunt series did not clearly delineate the extent of the pathology of the tuboovarian-shunt complex. Although ultrasound ruled out emergent surgical etiologies, such as a large ovarian cyst causing torsion and lack of perfusion to an ovary, or a tubal pregnancy, the patient did undergo multiple reevaluations until the true etiology was revealed under laparoscopy. Perhaps the layered mass involving several small structures fell beyond the specificity range of ultrasound.
Thus, an initial pelvic CT, not usually in our 1st-line arsenal for the adolescent patient with abdominal pain, might have eventually saved resources and decreased morbidity for this more complicated patient. A recent study using helical CT to differentiate appendicitis from acute gynecologic conditions in 100 females found the method to have 100 and 87% sensitivity, and 97 and 100% specificity, respectively, as well as high positive and negative predictive values. Helical CT findings resulted in a significant decrease in hospital admissions and an increase in accuracy of choice of correct surgical procedure (3). Another recent study employing CT as the diagnostic tool in 57 ED patients with acute nontraumatic abdominal pain found a 23.8% decrease in hospital admissions (4). One study used laparoscopy to successfully combine diagnosis and treatment for distal VP shunt malfunction in a series of 10 children (5).
Although the frequency of VP shunt malfunctions is cited up to 50% at 5 years, with the most common symptoms being headache, vomiting, and lethargy, abdominal findings are still described as rare (6). The most frequent types of malfunction continue to be mechanical obstruction and infection, but there are many interesting reports in the literature of less usual abdominal pathology. Summaries of the literature describe series of abdominal pseudocysts, as well as cases of intussusception, volvulus, incarcerated bowel, and perforation of bladder, stomach, colon, and gallbladder. Migration of the tubing into the mediastinum, umbilicus, and thoracic cavity has occurred. Perforation into the colon with anal protrusion and extraction as presumed ascaris worm, and gastric perforation with protrusion through the mouth have been reported. Genitourinary presentations are unique; intrascrotal migration with simulated testicular torsion and vaginal penetration (one child, one adult) have been reported (7). Migration to the fallopian tube has not been reported previously.
The cause of the abnormal shunt migration in this case was found to be more a factor of the patient’s postsurgical adhesions than shunt characteristics, since type of tubing and distal site choice were standard. VP shunts have become the preferred choice for pediatric patients, owing to the capacity of the peritoneal cavity to absorb fluid, ease of placement, and lack of thromboembolic events associated with vascular shunts (8). Other less typical sites of shunt placement have included the pleural cavity, the ureter, and gallbladder. One approach even described intentionally draining CSF into the fallopian tubes (9). Development of silastic (silicon) catheters, which did not carry the risk of distal plugging associated with polyethylene catheters, allowed a resurgence in the use of VP shunts in the 1970s (10). Rates of abnormal migration for each type of tubing are not known. A common cause for distal VP shunt migration in other cases has not been elucidated.
1. Leary FJ, Bass Jr, RB Symmondsa RE. Watery vaginal discharge in a young woman. J Urol 1979: 122:226–229.
2. Disaia PJ, Creasman WT. Clinical Gynecologic Oncology, ed 5. St. Louis: Mosby; 1997: 56–59, 375–376.
3. Rao PM, Feltmate CM, Rhea JT, et al. Helical computed tomography in differentiating appendicitis and acute gynecological conditions. Obstet Gynecol 1999; 93:417–421.
4. Rosen MP, Sands DZ, Longmaid 3rd, HE et al. Impact of abdominal CT on the management of patients presenting to the emergency department with acute abdominal pain. Am J Roentgenol 2000; 174:1391.
5. Esposito C, Porreca A, Gangemi M, et al. The use of laparoscopy in the diagnosis and treatment of abdominal complications of ventriculo-peritoneal shunts in children. Pediatr Surg Int 1998; 13:352–354.
6. Blount J, Campbell J, Haines S. Complications in ventricular cerebrospinal fluid shunting. Neurosurg Clin N Am 1993; 4:633–656.
7. Madsen MA. Emergency department management of ventriculoperitoneal cerebrospinal fluid shunts. Ann Emerg Med 1986; 15:1330–1343.
8. Lee TT, Uribe J, Ragheb J, et al. Unique clinical presentation of pediatric shunt malfunction. Pediatr Neurosurg 1999; 30:122–126.
9. Murtagh FR, Quencer RM, Poole CA. Extracranial complications of cerebrospinal fluid shunt function in childhood hydrocephalus. AJR Am J Roentgenol 1980: 135:763–766.
10. Couldwell WT, LeMay DR, McComb JG. Experience with use of extended length peritoneal shunt catheters. J Neurosurg 1996; 85:425–427.