Pediatr Infect Dis J, 2003;22:103942 Copyright 2003 by Lippincott Williams & Wilkins, Inc.

Vol. 22, No. 12 Printed in U.S.A.

Aseptic meningitis in infants younger than six months of age hospitalized with urinary tract infections
FELICE C. ADLER-SHOHET, MD, MICHELE M. CHEUNG, MD, MPH, MARYANN HILL, PHD AND JAY M. LIEBERMAN, MD

Background. Aseptic meningitis associated with urinary tract infection (UTI) in young infants has not been described in detail in the literature. We performed a retrospective study to determine the incidence and clinical features of aseptic meningitis accompanying UTI. Methods. We retrospectively reviewed the medical records of all infants younger than 6 months of age hospitalized with a UTI at Miller Childrens Hospital from March 1995 through March 2000. UTI was defined as a urine culture growing >10 000 colony-forming units/ml of a single organism from a catheterized specimen or >100 000 colony-forming units/ml of a single organism from a bagged urine specimen. Meningitis was defined as a positive cerebrospinal fluid culture or cerebrospinal fluid with >35 white blood cells/ mm3 in infants <30 days of age or with >10 white blood cells/mm3 in infants >30 days of age. Results. Of 386 infants with UTI, a lumbar puncture was performed in 260, and 31 (11.9%) had aseptic meningitis. One infant had bacterial meningitis. None of the 26 infants with UTI and bacteremia had aseptic meningitis. Two infants with meningitis had confirmed enteroviral infections, but aseptic meningitis did not occur more frequently in any particular month or during times of peak enteroviral activity. Conclusions. A cerebrospinal fluid pleocytosis is relatively common in hospitalized infants <6 months of age who have a UTI and usually does not reflect bacterial meningitis. Knowledge of this may prevent unnecessary courses of antibiotics for presumed bacterial meningitis and lead

to evaluation for other possible causes of aseptic meningitis including viral or congenital infections.
INTRODUCTION

Urinary tract infections (UTIs) are a common cause of fever in infants and frequently present as a febrile illness with no obvious source of infection. The diagnosis is often made after a complete evaluation for sepsis that includes urine, blood and cerebrospinal fluid (CSF) cultures. Blood cultures are positive in as many as 36% of infants with UTI, and bacteremia with UTI occurs almost exclusively in infants 6 months of age.1 4 Although concomitant bacterial meningitis is rare in infants with UTI, studies have noted young infants with UTI who have sterile CSF pleocytosis.2, 3, 5, 6 Most of these reports only briefly mentioned such infants and did not include evaluation for concomitant infections such as enteroviral meningitis, congenital infections or herpes simplex virus (HSV) meningoencephalitis. These reports also failed to discuss demographic or clinical factors possibly associated with meningitis. At our hospital the Infectious Diseases service is consulted several times a year to evaluate infants who have a UTI and sterile CSF pleocytosis. Some infants have been treated for up to 3 weeks with intravenous antibiotics for presumed bacterial meningitis despite negative CSF bacterial cultures. We performed a retrospective chart review of infants younger than 6 months of age hospitalized with UTI to evaluate the incidence and clinical features of aseptic meningitis in these infants.
METHODS

Accepted for publication Aug. 28, 2003. From Pediatric Infectious Diseases, Miller Childrens Hospital, Long Beach, CA (FCAS, MMC, JML); and University of California, Irvine, CA (FCAS, MMC, MAH, JML). Presented in part at the 39th annual meeting of the Infectious Diseases Society of America.17 Key words: Meningitis, urinary tract infection, infant. Address for reprints: Felice C. Adler-Shohet, M.D., Miller Childrens Hospital, 2801 Atlantic Ave., P.O. Box 1428, Long Beach, CA 90801-1428. Fax 562-997-9634; feliceadler@hotmail.com.

We reviewed hospital records of all children 6 months of age who were discharged from Miller Childrens Hospital from March 1995 through March 2000 with International Classification of Diseases, 9th revision (ICD-9) codes 590.00 to 590.9, 599.0, 047.0 to 047.9, 320.0 to 322.9, 049.1 and 053.0, corresponding to diagnoses of pyelonephritis, urinary tract infection and meningitis. For our analysis an infant was diagnosed as having a UTI when a urine culture grew 100 000 1039

1040

THE PEDIATRIC INFECTIOUS DISEASE JOURNAL

Vol. 22, No. 12, Dec. 2003

colony-forming units (CFU)/ml of a single organism from a bagged urine specimen or 10 000 CFU/ml of a single organism from a catheterized specimen. Patients whose urine cultures grew more than one organism were included only if they had urinary tract abnormalities or pyuria, which was defined as 10 white blood cells (WBC)/high powered field. Patients with hospitalacquired UTI, defined as a positive urine culture 24 h after hospital admission, were excluded. Meningitis was defined as a positive CSF culture or CSF pleocytosis. CSF pleocytosis was defined as 35 WBC/mm3 in infants 30 days of age and 10 WBC/ mm3 in infants 30 days.79 For traumatic lumbar punctures, CSF WBC counts were adjusted based on the ratio of white and red blood cells (RBC) in the peripheral blood. Renal ultrasound or voiding cystourethrogram examinations were considered abnormal when any gross renal anomalies, hydronephrosis, hydroureter, distension of the renal pelvises or vesicoureteral reflux were noted. The data were analyzed with SYSTAT statistical software. The two sample t test was used to compare differences in means and the chi square test of equal proportions was used for frequency tables. P 0.05 was considered significant. The study was approved by the Institutional Review Board at our hospital.
RESULTS

Review of hospital discharge ICD-9 codes identified 685 patients younger than 6 months of age discharged with a diagnosis of UTI or meningitis during the study period; 299 infants were excluded because they did not have a UTI or because they had a hospital-acquired UTI. The mean age of the 386 infants with UTI was 71.5 days (range, 2 to 182 days). Boys accounted for 77% of infants 0 to 30 days of age, 70% of infants 30 to 60 days of age, 60% of infants 61 to 90 days of age and 56% of infants 90 days of age (P 0.006). Urinary leukocyte esterase and nitrites were present in 83 and 33% of tested urine samples, respectively. Blood cultures were obtained from 353 infants (91%), of which 26 (7.4%) were positive. Male and female infants with a UTI were equally likely to have bacteremia (6.9% vs. 8.3%, respectively, P 0.6). Escherichia coli was the pathogen in 76% of infants with UTI alone, 96% of infants with UTI and bacteremia and 81% of infants with UTI and aseptic meningitis (P 0.06). Of the 386 infants with UTI, 260 (68%) had a lumbar puncture performed. Lumbar puncture was performed in 83 of 92 infants (90%) age 1 to 30 days, 85 of 104 infants (82%) age 31 to 60 days, 41 of 62 infants (66%) age 61 to 90 days and 51 of 128 infants (40%) age 91 days to 6 months. Thirty-one infants (11.9%) had

aseptic meningitis and one 7-week-old infant had bacterial meningitis, which was caused by E. coli. Laboratory features of the 31 infants with UTI and aseptic meningitis are shown in Table 1. The mean age of infants with UTI and aseptic meningitis was 71 days. In infants who had a lumbar puncture, the incidence of aseptic meningitis by age was 2.4% in infants younger than 31 days, 18.6% in infants 31 to 60 days, 9.8% in infants 61 to 90 days and 17.6% in infants 91 days to 6 months (P 0.006). Although boys accounted for 52% of infants with aseptic meningitis, girls with UTI were more likely than boys with UTI to have aseptic meningitis (17.6% vs. 9.1%, P 0.047). In infants with aseptic meningitis, the median CSF WBC count was 23 WBC/mm3 (range, 11 to 1180 WBC/mm3) with a median polymorphonuclear leukocyte (PMN) count of 4.6 cells/mm3 (range, 0 to 885 cells/mm3). In 90% of infants with aseptic meningitis, PMN accounted for fewer than one-half of total CSF WBC. All CSF Gram stains were negative for bacteria. Four infants with aseptic meningitis had received at least one dose of an antibiotic before lumbar puncture, and all four had normal CSF glucose and protein values with 45% PMN ( 15 PMN/mm3) in their CSF. Comparisons of clinical and laboratory features of infants with and without aseptic meningitis are shown in Table 2. Infants without aseptic meningitis had higher mean urinary WBC and were more likely to have urinary nitrites than infants with aseptic meningitis. Infants with UTI and aseptic meningitis were no more likely than infants with UTI alone to have abnormalities on renal ultrasound or voiding cystourethrogram. Studies to evaluate potential causes of aseptic meningitis were performed in 10 infants. These included CSF viral cultures (8 infants), CSF HSV polymerase chain reaction tests (2 infants), rectal and nasopharyngeal viral cultures (2 infants) and serum rapid plasma reagin (1 infant). Two infants with aseptic meningitis had positive enteroviral cultures, one from the CSF and the other from a rectal swab. Although more than one-half of the cases of enteroviral disease in our hospital occur between July and October, aseptic meningitis with UTI did not occur more frequently in any particular month or during times of peak enteroviral activity. There was no difference in duration of hospitalization between infants with and without aseptic meningitis (means of 4.7 days vs. 4.9 days, respectively; difference, 0.17; 95% confidence interval, 1.07 to 1.42; median, 4 days in both groups). However, two infants received 21-day courses of iv antibiotics for presumed bacterial meningitis despite negative CSF cultures. A 20-day-old infant had 1180 WBC in the CSF with 75% neutrophils. Because of his young age and suspicious CSF findings with negative CSF viral cultures, he

Vol. 22, No. 12, Dec. 2003

THE PEDIATRIC INFECTIOUS DISEASE JOURNAL

1041

TABLE 1. Laboratory features of 31 infants with urinary tract infection and aseptic meningitis
Urine Age (Days) 14 20* 31 31 35 35 36 36 38 41 45 47 48 50 51 55 59 60 61 61 81* 87 100 103 112 114 133 133 136 172 175 Sex WBC/hpf M M F F M M F M F M M F M M M F M M F M F F F F F F M M F F M 2 50 nd 25 nd 0 50 0 50 25 30 0 nd 40 5 12 3 5 50 7 50 1 5 50 40 50 50 50 50 9 35 Culture (colonies/ml) 50 000 100 000 Klebsiella 100 000 Escherichia coli 100 000 Enterococcus 100 000 E. coli 100 000 E. coli 100 000 Enterobacter 100 000 E. coli 10 000 50 000 Enterococcus 100 000 E. coli 100 000 E. coli 100 000 E. coli 10 000 50 000 E. coli 10 000 50 000 E. coli 100 000 E. coli 100 000 E. coli 100 000 E. coli 100 000 E. coli 10 000 50 000 Enterococcus 100 000 E. coli 100 000 E. coli 100 000 E. coli 10 000 50 000 Enterococcus 100 000 E. coli 100 000 E. coli 50 000 100 000 E. coli 100 000 E. coli 100 000 E. coli 100 000 E. coli 100 000 E. coli 100 000 E. coli 100 000 E. coli RBC/mm3 1093 240 3327 346 3930 6135 2 24 397 690 0 2 40 1 0 1649 101 4840 1 8 1 0 24 564 0 0 2 5 0 2428 0 Cerebrospinal Fluid WBC/mm3 55 1180 31 14 323 26 17 256 17 22 14 77 550 12 20 429 14 40 11 1118 28 341 14 15 208 13 23 13 690 19 13 PMN (%) 12 75 35 2 57 41 24 42 16 11 0 35 25 1 6 nd 2 37 0 3 3 17 20 34 20 2 0 10 2 67 1 Glucose (mg/dl) 46 40 53 48 49 76 66 45 56 44 47 48 42 49 47 53 64 69 63 52 66 50 74 63 75 68 66 46 52 nd 60 Protein (mg/dl) 105 106 91 87 77 34 42 109 73 75 72 47 99 59 37 88 44 42 27 52 33 69 20 31 15 22 24 26 39 nd 16

* Infant was treated with a 21-day course of intravenous antibiotics. Infant had positive culture for enterovirus. Infant received antibiotics before lumbar puncture. Urine was a bagged specimen. All other urine specimens from infants with aseptic meningitis were obtained by bladder catheterization. hpf, high power field; nd, not done.

received a 21-day course of iv antibiotics. The second infant had 28 WBC in the CSF. She was the daughter of a pediatrician who requested a prolonged antibiotic course for presumed culture-negative E. coli meningitis despite our recommendation to the contrary. This infant was hospitalized for 4 days and completed her iv antibiotic course at home. Comment. We found that concomitant aseptic meningitis was relatively common in infants younger than 6 months of age who were hospitalized with UTI, occurring in 12% of patients in whom lumbar puncture was performed. There are few previous reports of aseptic meningitis associated with UTI, and in some the association is mentioned only briefly.2, 3 In a short report of 1629 febrile infants 60 days of age or younger who were hospitalized in Israel and evaluated for sepsis, 11 had UTI and aseptic meningitis.5 However, neither diagnosis was clearly defined, and no further analyses were done. In a report from Greece, 15 (12.8%) of 117 infants younger than 90 days of age hospitalized with UTI had sterile CSF pleocytosis,6 similar to the incidence we found. In our study no clinical or laboratory features were predictive of aseptic meningitis. Although the presence of urinary nitrites and increased mean number of urine

WBCs were more common in infants without aseptic meningitis, these findings were not clinically useful for differentiating between groups. Aseptic meningitis occurred less frequently in infants younger than 1 month of age, although this may be a result of the more conservative definition of meningitis in younger infants. Female infants were more likely than male infants to have meningitis, but again this difference may be the result of our definition of meningitis, given that there was a higher percentage of female infants in the older age groups where meningitis was more prevalent. The length of hospitalization was not significantly different between infants with and without meningitis. In part this may reflect the fact that the Infectious Diseases consult service recommended against prolonged intravenous antibiotics in many of these infants. Enteroviral meningitis is more common during summer and early fall in southern California, but UTI with sterile CSF pleocytosis occurred throughout the year. Therefore it is unlikely that enteroviral infection was responsible for many cases of aseptic meningitis in this study. A more plausible hypothesis is that systemic inflammatory mediators or bacterial antigens such as lipo-

1042

THE PEDIATRIC INFECTIOUS DISEASE JOURNAL

Vol. 22, No. 12, Dec. 2003

TABLE 2. Clinical and laboratory information in infants with UTI vs. infants with UTI and aseptic meningitis
UTI (N 355) 71.5 67 84 232/276 35 103/298 51 (310) 38 (119/314) 23 34/149 8 26/324 27 4.9 UTI with Aseptic Meningitis (N 31) 71 52 72 18/25 15 4/26 27 (28) 26 (7/27) 18 2/11 0 0/29 32 4.7

Clinical/Laboratory Information

Mean age (days) Male (%) Urine leukocyte esterase present % N/total Urine nitrites present* % N/total Mean urine WBC/hpf N Abnormal renal ultrasound % N Abnormal VCUG % N/total Bacteremia % N % presenting during enteroviral season (JulyOctober) Mean days of hospitalization

CSF results can be difficult, only 2 infants had 4000 RBC in the CSF. Although clinicians may disagree with the inclusion of some of these patients, even if all infants with 1000 RBC in their CSF were excluded, 9% of hospitalized infants 6 months of age with UTI had aseptic meningitis. We conclude that for infants with UTI and sterile CSF pleocytosis, treatment should be directed at the UTI unless compelling evidence exists for bacterial meningitis or another central nervous system infection.
REFERENCES
1. Ginsburg CM, McCracken GH Jr. Urinary tract infections in young infants. Pediatrics 1982;69:409 12. 2. Bachur R, Caputo GL. Bacteremia and meningitis among infants with urinary tract infections. Pediatr Emerg Care 1995;11:280 4. 3. Bergstrom T, Larson H, Lincoln K, Winberg J. Studies of urinary tract infections in infancy and childhood: XII. Eighty consecutive patients with neonatal infection. J Pediatr 1972; 80:858 66. 4. Wiswell TE, Geschke DW. Risks from circumcision during the first month of life compared with those for uncircumcised boys. Pediatrics 1989;83:101115. 5. Finkelstein Y, Mosseri R, Garty BZ. Concomitant aseptic meningitis and bacterial urinary tract infection in young febrile infants. Pediatr Infect Dis J 2001;20:630 2. 6. Syrogiannopoulos GA, Grivea IN, Anastassiou ED, et al. Sterile cerebrospinal fluid pleocytosis in young infants with urinary tract infection. Pediatr Infect Dis J 2001;20:92730. 7. Ahmed A, Hickley SM, Ehrett S, et al. Cerebrospinal fluid values in the term neonate. Pediatr Infect Dis J 1996;15:298 303. 8. Bonadio WA. The cerebrospinal fluid: physiologic aspects and alterations associated with bacterial meningitis. Pediatr Infect Dis J 1992;11:42332. 9. Bonadio WA, Stanco L, Bruce R, Barry D, Smith D. Reference values of normal cerebrospinal fluid composition in infants ages 0 to 8 weeks. Pediatr Infect Dis J 1992;11:589 91. 10. Tullus K, Fituri O, Burman LG, Wretlind B, Brauner A. Interleukin-6 and interleukin-8 in the urine of children with acute pyelonephritis. Pediatr Nephrol 1994;8:280 4. 11. Benson M, Jodal U, Agace W, et al. Interleukin (IL)-6 and IL-8 in children with febrile urinary tract infection and asymptomatic bacteriuria. J Infect Dis 1996;174:1080 4. 12. Kassir K, Vargas-Shiraishi O, Zaldivar F, et al. Cytokine profiles of pediatric patients treated with antibiotics for pyelonephritis: potential therapeutic impact. Clin Diagn Lab Immunol 2001;8:1060 3. 13. Jacobson SH, Hylander B, Wretlind B, Brauner A. Interleukin-6 and interleukin-8 in serum and urine in patients with acute pyelonephritis in relation to bacterial-virulenceassociated traits and renal function. Nephron 1994;67:1729. 14. Chavanet P, Bonnotte B, Guiguet M, et al. High concentrations of intrathecal interleukin-6 in human bacterial and nonbacterial meningitis. J Infect Dis 1992;166:428 31. 15. Dulkerian SJ, Kilpatrick L, Costarino AT, et al. Cytokine elevations in infants with bacterial and aseptic meningitis. J Pediatr 1995;126:872 6. 16. Lopez-Cortes LF, Cruz-Ruiz M, Gomez-Mateos J, et al. Interleukin-8 in cerebrospinal fluid from patients with meningitis of different etiologies: its possible role as neutrophil chemotactic factor. J Infect Dis 1995;172:581 4. 17. Adler-Shohet FG, Cheung MM, Hill M, Lieberman JM. Aseptic meningitis in infants with urinary tract infections [Abstract 295]. In: 39th annual meeting of the Infectious Diseases Society of America, San Francisco, October 25 to 28, 2001. Clin Infect Dis 2001;33:1139.

* P 0.046. P 0.006. N, number of patients; hpf, high power field; VCUG, voiding cystourethrogram.

polysaccharides released as a consequence of pyelonephritis lead to sterile pleocytosis. Lipopolysaccharides induce a cytokine cascade that results in cytokine release systemically and perhaps in the CSF as well. Several studies have measured urine and/or serum interleukin-6 (IL-6) and interleukin-8 (IL-8) in individuals with febrile UTI. Children with febrile UTIs have higher urinary concentrations of IL-6 and IL-8 than children without UTIs.10 12 In addition serum concentrations of these two cytokines were higher in women13 and children11 with pyelonephritis than in controls. IL-6 and IL-8 have been found in the CSF of patients with meningitis14 16 and are thought to play a role in the initiation of the CSF pleocytosis that occurs during meningitis. This retrospective review has a number of limitations. Most infants with meningitis had no further evaluation; therefore it is possible that cases of enteroviral infection or congenital infection were missed. It is unlikely that any cases of HSV were missed, given that this infection would have progressed without appropriate therapy. Not all infants with UTI were the subjects of blood culture and lumbar puncture; thus the true incidence of CSF pleocytosis accompanying UTI cannot be defined. Infants who were more ill appearing might have been more likely to have these tests, which could have biased the results. Positive urine cultures from bagged urine specimens were included, but only if the urine grew 100 000 CFU of a single organism per ml. This was the case for 17 infants, only 1 of whom had aseptic meningitis. Traumatic lumbar punctures were also included, and although interpretation of these