Epidural Abscess in a One-Year-Old Boy After Continuous Epidural Analgesia

Regional anesthetic techniques have become increasingly popular in the management of perioperative and postoperative pain in children. In a recent study, Flandin-Blety and Barrier [1] presented a large series of children with complications after extradural analgesia. Among these, five cases with neurological sequelae were described. Strafford et al. [2] reported 1620 children treated postoperative with short-term (0-8 days) epidural analgesia. They concluded that the risk of epidural infections was quite low in pediatric patients receiving short-term catheterization. Emmanuel [3] described an abscess in a 4-year-old child that developed after placement of an extradural catheter in the sacral region and responded satisfactorily to antibiotics. There are only a few studies concerning the long-term use of lumbar epidural catheters, especially in children [4]. This report describes a one-year-old boy who developed an epidural abscess after he had been receiving epidural analgesia for 12 days. This case is remarkable for the benign presentation of the epidural abscess, its extensiveness, and its successful nonsurgical treatment over 11 days. Case Report A 5-mo-old boy was admitted to St. Goran's Children's Hospital because of severe visceral pain. Chronic intestinal pseudoobstruction was diagnosed [5]. In the present case, the syndrome consisted of episodes of severe abdominal pain, sometimes combined with septicemia. The episodes of pain were characterized by crying, increased muscle tone, grayish color of the skin, hypertension, tachyarrhythmia, and high fever (40 degrees C). They varied in frequency, severity, and duration. The pain was so intense that intravenous (IV) morphine had to be administered. If treatment were not sufficient, the episodes of pain could persist for several hours. Initially, the patient had pain-free intervals, and only intermittent treatment was necessary. At 7 mo of age, mechanical obstruction supervened, and a colostomy was performed for decompression. Treatment of the postoperative pain was started with a lumbar epidural infusion using a local anesthetic, bupivacaine 1.25 mg/mL (0.4 mg [centered dot] kg-1 [centered dot] h-1), maintained for 3 days. The analgesic effect was satisfactory, and no complications developed. Six months later (at 1 yr of age), the patient was readmitted to the hospital because of a recurrence of the severe episodes of the abdominal pain (which had the same characteristics as before). He was admitted to the pediatric intensive care unit, and an infusion of a morphine (22 micro g [centered dot] kg-1 [centered dot] h-1) was started. However, it did not alleviate his intense pain sufficiently, even when combined with IV bolus doses of 50-150 micro g/kg. Apart from opioids, the following drugs were administered to treat the attacks of pain: adenosine, clonidine, cisaprid, ketamine, lidocaine (I.V.), ondansetron, phenytoin, and propacetamol. After discussing and evaluating the patient's situation with the physicians involved, it was decided that surgical intervention would not be beneficial and that alternative methods for treating the pain should be tried. A lumbar epidural catheter was therefore inserted. The entry site was covered with a thin, transparent plastic film (Tegaderm[TM]; 3M Canada, Ontario, Canada). According to hospital routine, the catheter entry site was inspected daily for evidence of swelling, tenderness, and infection. A combination of local anesthetics (bupivacaine 1.25 mg/mL, 0.4 mg [centered dot] kg-1 [centered dot] h-1) and epidural opioids (morphine 33 micro g/kg x 3) was used, with good initial effect. Seven days later, the analgesic effect diminished, although the doses had been increased. The loss of analgesic effect was ascribed to tachyphylaxis in combination with a less than optimal placement of the tip of the catheter. A radiogram showed that the tip of the lumbar epidural catheter was located at the level of L4. The patient's situation was reevaluated, and it was decided that surgical intervention was not indicated at that time. A multidisciplinary decision to replace the epidural catheter was made, taking the risk of infection into consideration. Antibiotic treatment with vancomycin and imipenen was started at the same time due to concerns about septicemia. The child was believed to have terminal disease, and the analgesic treatment was initiated on this basis. Subsequently, the lumbar catheter was replaced by a catheter placed at the level of T11-12, its position being confirmed on a radiogram. An epidural infusion was started, using a combination of bupivacaine 1.25 mg/mL (0.4 mg [centered dot] kg-1 [centered dot] h-1), fentanyl (2 micro g/mL), and clonidine (3.3 micro g/kg x 3). At first, this combination had a good analgesic effect, but 3 days later, supplementary doses of IV morphine were required. Eleven days after placement of the third epidural catheter, the concentration of the local anesthetic had to be increased to 3.75 mg/mL. The patient was then receiving 1.1 mg [centered dot] kg-1 [centered dot] h-1 of bupivacaine. Despite the large dose infused, the pain was not reduced, and supplementary doses of opioids were needed. No clinical side effects from the large doses of local anesthetic were seen. Treatment with the antibiotics continued. Twelve days after placement of the third epidural catheter, tender swelling was detected at the penetration site of the epidural catheter (L1-2). The catheter was immediately removed. The epidural catheter had been in place for 20 days. A computed to tomography (CAT) scan revealed an intraspinal process in the lumbar and thoracic areas. Magnetic resonance imaging (MRI) confirmed the presence of an edpidural abscess (Figure 1). The process extended from T5 to L5, deforming and dislocating the medulla in this area. Surprisingly, the patient developed no neurological symptoms.Figure 1: Magnetic resonance imaging of the epidural abscess. The process extended from T5 to L5.Antibiotic therapy with cefuroxim was initiated (the same day that the previous antibiotic therapy had been terminated to establish an antibiotic-free interval to enable blood cultures to be taken from the patient). A bacterial culture from the epidural catheter tip revealed the growth of Pseudomonas aeroginosa. A CAT scan-aided puncture of the epidural process was performed, and the bacterial culture confirmed the growth of the same bacteria. There was no indication for neurosurgical intervention. After removal of the catheter, it was established that the bacteria was sensitive to tobramycin and only partly sensitive to cefuroxim. Subsequently, the antibiotic therapy was changed. There was no obvious evidence of septicemia when the third epidural catheter was installed, although antibiotic treatment with vancomycin and imipenen was started due to suspected septicemia. The white cell count was not increased, but the cross-reactive protein did show an increase from baseline to 100 the day before the abscess was discovered. The level of cross-reactive protein peaked at 147 mg/L 6 days after removal of the catheter, and it decreased during the following days. Five days after the epidural abscess was detected, another CAT scan showed that the size of the abscess had decreased substantially, and after 11 days, the abscess could no longer be seen. After epidural catheter removal, an infusion of morphine was started, and large doses (400 micro g [centered dot] kg-1 [centered dot] h-1) were required to relieve pain adequately. The characteristics of the pain had not changed. Large supplementary doses of morphine (1-5 mg/kg) were often needed. No clinical side effects were noted. The patient's general condition deteriorated, and it was decided to perform a colectomy because of this life-threatening condition. Thereafter, the frequency and severity of the colic pain diminished, and the patient recovered remarkably well. The child has now been accepted for transplantation of the intestines. Discussion Schuffler and Jonak [6] described chronic intestinal pseudoobstruction as a syndrome that affects the regulation of intestinal propulsion, whether it be of muscular, neurogenic, or hormonal origin. The etiology is not known. This syndrome is sometimes associated with abdominal pain of varying intensity [7]. Chronic intestinal pseudoobstruction usually leads to death before the age of one year [8]. In this case, the intensity of the pain often peaked when the patient developed septicemia. The origin of the pain is not known, but it could be the net result of an autonomic imbalance causing gastrointestinal atony, bacterial proliferation, and degeneration of the intestines. Consequently, the pain may be of neurogenic origin, in combination with ischemic nociceptive pain. The patient's pain gradually worsened in intensity as well as in frequency. A chronic state of pain developed, probably due to the underlying intestinal disease. IV opioids had little effect when given in the usual doses (50-150 micro g/kg), but large doses (1-3 mg/kg) alleviated the abdominal pain. The abscess was probably caused by contamination of the epidural catheter as the end result of continuous bacteremia from the necrotizing intestines. The bacteria found in the abscess (P. aeroginosa) had previously been isolated in blood cultures and in the region of the iliostomy. The rapid development of tachyphylaxis may have been due not only to tolerance to the local anesthetic but also to the development of an epidural abscess and its gradual enlargement. The abscess might limit the spread of the epidural infusion and thereby reduce the analgesic effect. In retrospect, when the effect of the epidural analgesia diminished after seven days, an epidural abscess should have been included in the differential diagnosis. Surprisingly, no neurological symptoms were present, although the epidural abscess compromised the medulla over several segments. No sequelae were seen after two months. The present case report illustrates an unusually difficult pain condition that only responded to very large doses of opioids. Epidural analgesia was effective, but it entailed some risks. In summary, a one-year-old boy with chronic intestinal pseudoobstruction was treated in the pediatric intensive care unit at St. Goran's Children's Hospital because of severe visceral pain. He received epidural analgesia three times: the first time as postoperative pain management after laparotomy because of mechanical obstruction; the second and third time the patient was believed to be terminally ill, and the treatment of the pain was initiated on this basis. Twelve days after placement of the third epidural catheter, the development of an epidural abscess was suspected. An MRI confirmed the presence of an epidural abscess extending from T5 to L5. Surprisingly, the patient developed no neurological symptoms. Treatment of the abscess was managed nonsurgically, and antibiotic therapy was immediately initiated. After 11 days, the abscess could no longer be detected with MRI. No sequelae was seen. Epidural analgesia is an established and well documented method for treating pain in children [9]. The risk of infections seems to be low [2,4]. In patients who are susceptible to infections, one must weight the value of such therapy against the risk. This was an extreme situation, requiring extreme measures.