User login
Scenario: A 32-day-old female presents with vomiting and jaundice, and the mother reports that the child is irritable and not eating as well as usual. The pediatric hospitalist suspects a UTI and orders a dipstick/urinalysis and urine culture. The dipstick/urinalysis results are positive for nitrite and leukocyte esterase, indicating antibiotic therapy for presumed UTI.
Introduction
UTI is one of the most common bacterial infections in infants and young children. To prevent progression to pyelonephritis and avoid potential renal scarring or failure, early recognition and prompt treatment are critical.
Clinical signs and symptoms of UTI in newborns include jaundice, sepsis, failure to thrive, poor feeding, vomiting, and fever. In infants and preschoolers, hospitalists should also suspect UTI in the presence of diarrhea, strong-smelling urine, abdominal or flank pain, and new onset urinary incontinence.1
Treatment recommendations, which are age-dependent, include antibiotic therapy initiated upon an abnormal dipstick/urinalysis. If a urine culture is positive, a seven- to 14-day course of antibiotic therapy is recommended, followed by prophylactic antibiotics until results of imaging studies are available.1 According to a study by Hoberman and Wald, treatment of UTI with oral antibiotics alone is generally effective, even for young children with pyelonephritis.2
Imaging recommendations for a first UTI include ultrasound, cystogram, and renal cortical scan.
Up to this point, there’s been very little disagreement about the management of UTI. However, the question of whether the one-month-old patient in our scenario should be admitted or sent home with strict instructions on the administration of antibiotics remains controversial. The pediatric UTI guideline from Cincinnati Children’s Hospital Medical Center recommends routine hospital admission if the infant is under 30 days old.1 On the other hand, Santen and Altieri, among others, recommend, “Sick children and infants less than three months should be treated as inpatients, and healthy children and older infants may be treated as outpatients.”3,4
—Jeffrey M. Simmons, MD
The Set-Up
We asked several pediatric hospitalists across the country—including an instructor of pediatrics at Cincinnati Children’s Hospital Medical Center—to respond to this simple scenario, posing the following questions: “What recommendation would you follow? Would you admit the one-month-old infant in our scenario or send her home? Why?”
Predictably, admission criteria varied, but most agreed that this infant should be admitted. No one based their response on age.
Automatic Admission Cut-off Not Well Supported
Jeffrey M. Simmons, MD, instructor of pediatrics, Division of General and Community Pediatrics, Cincinnati Children’s Hospital Medical Center, responds: “My perspective on the UTI scenario is that the only dogma that applies is that the infant needs antibiotics. Because [the scenario] mentions vomiting, I would, therefore, most likely give parenteral antibiotics and admit.
“An issue within this scenario that I don’t believe the literature answers clearly is once a UTI is identified by the U/A, what risk remains of bacteremia or meningitis in the over 30-day-old infant? We are taught that infants don’t ‘localize’ infections well (i.e., a serious bacterial infection in one place can rapidly lead to disseminated infection). I sense a growing consensus that after 30 days or so this concern is less at issue. However, for clarity, I would prefer to obtain blood and CSF cultures on this infant prior to initiating antibiotic therapy. Without those cultures, and the screening tests that go with them (i.e., serum WBC count, CSF cell count, and glucose), I would be uncomfortable sending the infant home.
“However, if the WBC count was between 5 and 15, the CSF reassuring, the parents in agreement and reliable with good primary care follow-up the next morning, I think such an infant could be managed with one dose of IV/IM antibiotics after cultures are obtained and sent home. The following day, depending on culture results and the clinical situation, this infant might either be admitted, given another dose of parenteral antibiotics pending final blood culture, or switched to oral antibiotics.
“Such a plan is complex, so if the primary care physician or family was at all uncomfortable, admission to accomplish that plan would clearly be appropriate.
“A final issue not adequately addressed within the scenario is an assessment of the infant’s hydration—potentially an issue due to the vomiting. If the infant was mildly to moderately dehydrated with persistent vomiting, I would also then admit until this improved.
“In regard to the cited evidence, I would say that clearly an arbitrary cut-off of automatic admission under three months is not well supported by current literature, but is certainly many practitioners’ ‘style.’ Oral antibiotics are clearly efficacious for pyelonephritis, but the clinician needs to be confident the family can give them and the infant will keep them down. I believe most practitioners would agree with admission for an infant under 30 days, but where to precisely draw that line needs to be better established through investigation. Potential concomitant bacteremia and meningitis lead me to support obtaining blood and CSF cultures on any infant under 60 days for which I plan to initiate antibiotics for UTI.”
Possibility of Dehydration Indicates Admission
Michael P. Fullmer, DO, Central Iowa Pediatric Hospitalists, Mercy Medical Center, Des Moines, Iowa, responds: “This patient is not febrile, but does have other systemic symptoms, including poor feeding, irritability, jaundice, and vomiting. These systemic symptoms could be indicative of a serious bacterial infection like UTI, bacteremia, or meningitis. This patient most likely has a UTI. These symptoms may suggest pyelonephritis rather than lower UTI, but the distinction is not necessary for our decision here.
“Bacteremia is present in up to 22.7% of infants less than two months old with a UTI. This adds to the complexity of the issue, and a blood culture is probably indicated. Meningitis should be considered, but a lumbar puncture is probably not indicated in this scenario.
“A one-month-old with a UTI should be admitted to the hospital and started on parenteral antibiotics. There are several reasons for admission. First, the presence of vomiting makes oral antibiotic administration impractical (if not impossible). Intramuscular antibiotic injection may be an option, but the IV route gives the provider more options and is usually better accepted by parents. Next, if the patient is not already dehydrated, the poor feeding and vomiting could lead to dehydration. This alone would be an indication for admission. Finally, admission would give the physician time to observe the infant for clinical improvement. This may be subtle in the absence of fever.
“Another consideration is the recommended imaging for all children less than two years old. The AAP has recommended a renal ultrasound and VCUG for all infants and young children with their first UTI. This is important for discovery of urinary tract anomalies that predispose the patient to recurrent UTIs and eventual renal scarring and dysfunction. The imaging should be performed at the earliest convenient time if the patient is responding to therapy. In our practice, we generally have the renal ultrasound performed while the patient is in the hospital and arrange the VCUG as an outpatient [exam] prior to completion of the course of antibiotics.
“Patients are discharged when they are afebrile for 24 hours, have adequate oral intake and are able to take oral antibiotics. Please refer to the AAP Clinical Practice Guideline for more detail.5,6
—Michael Fullmer, DO
Admission Criteria Must Take Fever Into Account
John W. Graef, MD, chief, Services Office at Children’s Hospital, Harvard Vanguard Services Office, Boston, responds: “The presence or absence of fever is an important variable. Dipstick urines are shortcuts and don’t provide such information as the presence or absence of casts, although the fact that the child is vomiting and irritable is suggestive of pyelonephritis as opposed to a simple UTI/cystitis. [The scenario doesn’t] mention how the urine is obtained. Presumably it is a cath specimen, but that needs to be specified.
“Jaundice can occur with a UTI, but usually in the first week or so of life. A 32-day-old infant with pyelo is unlikely to be jaundiced unless for some other reason.
“I certainly agree with routine admission of a febrile infant up to one month, but not necessarily an otherwise well female infant. The presence of vomiting and irritability with or without fever might prompt a full septic work-up, in which case the decision to admit an afebrile infant would depend on the results of the CBC/UA and probably an LP. A blood culture and CBC should be drawn in the ED.
“In other words, one can’t have it both ways. If the irritability and vomiting are due to pyelo, a septic work-up is warranted. If all parameters are normal, the only reason for admission is hydration of a vomiting infant. If the infant was afebrile and had an uncomplicated UTI, I would not automatically admit an otherwise well 32-day-old.
“Poor PO (per OS, i.e., oral, by mouth) intake in an infant with a UTI warrants IV fluid regardless of age.”
—John W. Graef, MD
True Emesis With Decreased Oral Intake Indicates Admission
Erin R. Stucky, MD, pediatric hospitalist, Children’s Hospital and Health Center San Diego, associate professor, UCSD Pediatrics, responds: “The decision to admit or discharge this one-month-old with vomiting, jaundice, poor eating, and irritability should include consideration of hydration status, toxicity, sepsis potential, and ability to secure close follow-up. We will make the assumption that the infant is term, with no past medical history, family history, or prenatal evaluation that would put the infant at greater risk for sepsis or likelihood of underlying anatomic genitourinary abnormality. In addition we will assume that the urinalysis was performed by catheterization in a non-pretreated infant.
“A careful history should elicit the change in urine output, frequency, and volume of emesis to contrast with small ‘spit-ups,’ and change in feeding duration or volume. Observation of a feeding in the office or emergency department can be of great value in determining likelihood of ability to maintain hydration at home. Feeding type should be confirmed [because] parents may dilute or alternately concentrate formula in response to vomiting. In this scenario, true emesis in an infant with decreased oral intake would be grounds for admission, intravenous hydration, and evaluation of electrolytes.
“A newly irritable infant evokes a visceral sensation for all pediatric hospitalists. An irritable one-month-old may be suffering from a single system infection, electrolyte imbalance, or other insult, but unfortunately may just as easily have multisystem involvement. Vital signs and physical exam findings of toxicity, such as tachycardia and delayed capillary refill, are not known to us. The presence of jaundice raises concern for cholestasis induced by E. coli or possibly rarer metabolic disease, such as galactosemia. Admission would allow for evaluation and monitoring of the more likely causes of irritability in our index patient, which include sepsis, meningitis, and electrolyte and acid-base imbalances.
“The urinalysis in this infant is suggestive of a urinary tract infection, although infants may have no abnormalities noted on initial urinalysis.7,8 The risk of bacteremia in infants under 60 days with documented urinary tract infection is significant. A number of studies support the need to treat infants less than 30 days with parenteral antibiotics.9-14 Addition of C-reactive protein testing at this time does not aid in distinguishing those who are bacteremic in this age group.15 The often quoted study by Hoberman of 306 children included only 13 under the age of two months.16 Of the 13 reported positive blood cultures, 10 were in children under age six months. Daily intramuscular ceftriaxone treatment would cover a majority of the typical neonatal UTI organisms, can be administered in the outpatient setting, and is proven to be as effective as intravenous delivery. The clinical response to bacteremia is, however, unpredictable in young infants. The sepsis potential in this infant requires admission for physiologic monitoring and support as needed.
“Final, but not inconsequential, concerns are barriers to follow-up. These include parental experience and coping skills with feeding and monitoring an ill infant, ability to educate on the illness and reasons for follow-up, transportation, and operational issues, such as weekend clinic hours or holiday office closures. For the index patient these issues are overshadowed by the clinical criteria for admission but would be of great importance for discharge.”
Conclusion
Based on these responses, admitting a suspected UTI patient on the basis of age alone, as suggested by Santen and Altieri, is likely inappropriate. Many other factors must be weighed and would likely indicate admission for the patient in the scenario, regardless of the infant’s age. In short, until there’s better evidence for age-based admission criteria, clinical judgment based on the individual patient presentation must continue to drive care and treatment decisions.
Keri Losavio is a medical journalist with more than 10 years’ experience.
References
- UTI Guideline Team, Cincinnati Children’s Hospital Medical Center. “Evidence based clinical practice guideline for children 12 years of age or less with acute first time urinary tract infection.” www.cincinnatichildrens.org/svc/dept-div/health-policy/ev-based/uti.htm. Guideline 7, pages 1–20, April 2005.
- Hoberman A, Wald ER. Treatment of urinary tract infections. Pediatr Infect Dis J. 1999;18(11):1020–1021.
- Santen SA, Altieri MF. Pediatric urinary tract infection. Emerg Med Clin North Am. 200119(3):675–690.
- Egland AG, Egland TK. Pyelonephritis. eMedicine. www.emedicine.com/emerg/topic769.htm. Accessed Oct. 16, 2005.
- No authors listed. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Pediatrics. 1999;103(4):843–852.
- Pitetti RD, Choi S. Utility of blood cultures in febrile children with UTI. Am J Emerg Med. 2002;20:271–274.
- Dayan PS, Bennett J, Best R, et al. Test characteristics of the urine Gram stain in infants 60 days of age with fever. Pediatr Emerg Care. 2002;18(1):12–14.
- Huicho L, Campos-Sanchez M, Alamo C. Meta-analysis of urine screening tests for determining the risk of urinary tract infection in children. Pediatr Infect Dis J. 2002;21 (1):1-11.
- Byington C L, Enriquez F, Hoff C, et al. Serious bacterial infections in febrile infants 1 to 90 days old with and without viral infections. Pediatrics. 2004; 113(6):1662–1666.
- Baraff L. Management of fever without source in infants and children. Ann Emerg Med. 2000;36(6):602–614.
- Baraff LJ, Oslund SA, Schriger DL, Stephen ML. Probability of bacterial infections in febrile infants less than three months of age: A meta-analysis. Pediatr Infect Dis J. 1992;11(4):257–264.
- Klein JO. Management of the febrile child without a focus of infection in the era of universal pneumococcal immunization. Pediatr Infect Dis J. 2002;21(6):584–588.
- Syrogiannopoulos G, Grieva I, Anastassiou E, et al. Sterile cerebrospinal fluid pleocytosis in young infants with urinary tract infections. Pediatr Infect Dis J. 2001;20(10):927–930.
- Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febrile infants at low risk for serious bacterial infection—an appraisal of the Rochester criteria and implications for management. Febrile Collaborative Study Group. Pediatrics. 1994;94(3):390–396.
- Malik A, Hui C, Pennie RA, Kirpalani H. Beyond the complete blood cell count and C-reactive protein: A systematic review of modern diagnostic tests for neonatal sepsis. Arch Pediatr Adolesc Med. 2003;157(6):511–516.
- Hoberman A, Wald ER, Hickey RW, et al. Oral versus intravenous therapy for urinary tract Infections in young children. Pediatrics.1999;104:79–86
PEDIATRIC SPECIAL SECTION
IN THE LITERATURE
Utilize Clinical and Demographic Factors to Diagnose UTIs in Young Febrile Infants
Review by Sara E. Gardner, MD
Zorc JJ, Levine DA, Platt SL, et al. Clinical and demographic factors associated with urinary tract infections in young febrile infants. Pediatrics. 2000;116(3):644-648.
UTI is a common cause of serious bacterial infection in the febrile infant <60 days of age. Standard urinalysis and urine dipstick techniques, commonly used to diagnose UTI, have relatively low sensitivity increasing the possibility of a missed diagnosis. An accurate initial diagnosis is critical in this age group for whom complications from UTI include bacteremia and renal scarring.
To describe the demographic and clinical factors associated with UTI in infants ≤60 days of age with fever, these authors conducted a prospective cross sectional study from October 1999 to March 2001. Patients were enrolled at eight different institutions after presentation to an emergency department. One-thousand-twenty-five patients age 60 days or younger (mean age 35.5 days, 60.5% male) were enrolled with either reported or documented fever >38. Routine testing for all children included respiratory syncytial virus (RSV) sampling and bladder catheterization or suprapubic aspiration for urinalysis and culture.
A positive urinalysis was defined as a trace or greater result for leukocyte esterase and/or nitrite on dipstick or greater than or equal to five WBCs per high power field (hpf) on urine microscopy. UTI was defined as growth of a single pathogen of ≥1,000 colony forming units (cfu)/mL for urine cultures obtained by suprapubic catheterization, ≥50,000 cfu/mL from a catheterized specimen, or ≥10,000 cfu/ml from a catheterized specimen with a positive urinalysis.
Of the patients enrolled in the study, 92 were found to have UTI by these diagnostic criteria. Using the chi-squared test and calculated odds ratios with 95% confidence intervals, uncircumcised male (OR: 10.4; 95% CI: 4.7-31.4) and maximum temperature of ≥39º C (OR: 2.4 per degree C; 95% CI: 1.5-3.6) were found to be statistically significant variables for predicting UTI. These risk factors remained statistically significant after multivariable analysis controlling for other factors.
Interestingly of the above 92 patients diagnosed with UTI, 85 grew ≥50,000 cfu of a single pathogen, but six (8%) grew 10,000-49,000 cfu with a positive UA based on the study criteria. Zorc, et al. included these patients with >10,000 cfu and >5 WBC/hpf in this study despite previous studies that have established a definition of positive urinalysis to be ≥10 WBC/hpf. Zorc, et al. acknowledge the conservative definition applied in their current study, but assert that the overall results of the study would have been similar had 10,000 or 50,000 cfu/mL thresholds been chosen. To support this assertion, Zorc, et al. retrospectively applied enhanced urinalysis, a sensitive form of urinalysis including hemocytometric cell count and gram stain described by Hoberman, et al. to study patients with low bacteria counts. Based on Hoberman’s study, enhanced urinalysis can differentiate acute infection from asymptomatic bacteriuria in patients with bacterial growth between 10,000 to 50,000 cfu/mL.
Another significant limitation of this study was failure to enroll one-third of eligible patients to the study. In addition, the authors note that missed patients had a lower rate of UTI compared with enrolled patients.
Although this study design prohibits generalization to patient care areas outside the emergency department, the findings can assist the hospitalist in the evaluation of the febrile infant during RSV season and potentially guide decisions regarding empiric antibiotic therapy as part of evaluations to diagnose or exclude serious bacterial infection. Additionally, this study raises the question of need for better methods of urinalysis for febrile, uncircumcised male infants.
Back to the Basics: A Clinical Approach to Excluding Acute Appendicitis
Review by Jane G. Buss, MD
Kharbanda AB, Taylor GA, Fishman SJ, et al. A clinical decision rule to identify children at low risk for appendicitis. Pediatrics. 2005;116(3):709-16.
Appendicitis is the most common condition requiring emergency abdominal surgery in the pediatric population. To diagnosis appendicitis, clinicians typically utilize clinical findings, imaging studies, and laboratory testing. The use of clinical scoring systems to improve the diagnosis of appendicitis in children has been limited, mostly resulting from unacceptably low sensitivities/specificities, or the lack of validation. The use of CT scan to aid in the evaluation of children with appendicitis has become widespread. Concerns remain, however, regarding long-term radiation effects and increased healthcare costs associated with this approach.
The authors of this study sought to develop clinical scores for children to predict which of those with acute abdominal pain do not have appendicitis. Additionally, they hoped to lessen the use of CT scanning. Their goal was to identify those groups of children who have significant abdominal pain without appendicitis who could be safely observed without CT scan or possible surgery.
This prospective cohort study enrolled 601 eligible children ages three to 18 with suspected appendicitis who presented to the emergency department over a 15-month period. Two low-risk clinical decision rules were developed using logistic regression and recursive partitioning. Using logistic regression, six factors were identified from analysis of 425 patients in the derivation set significantly associated with an increased likelihood of appendicitis. The score components include:
- Nausea (2 points);
- History of focal right lower quadrant pain (RLQ) (2 points);
- Migration of pain (1 point);
- Difficulty walking (1 point);
- Rebound tenderness (2 points); and
- Absolute neutrophil count (ANC) >6.75 x 103/uL (6 points).
When tested with data from the 176 patients in the validation set, a score of less than or equal to five had a sensitivity of 96.3%, and a negative predictive value of 95.6% for excluding appendicitis.
The authors then derived a second clinical low-risk decision rule by recursive partitioning. They determined that a combination of ANC <6.75 X 10 to the third/microliter, absence of nausea (or emesis or anorexia), and absence of maximal tenderness in the RLQ essentially excluded appendicitis in the derivation and validation groups. This rule had a sensitivity of 98.1%, and a negative predictive value of 97.5%.
In summary, these authors derived two clinical decision rules giving the clinician the option of a clinical score (logistic regression) or a decision tree (recursive partitioning) to identify children at low risk for appendicitis. Their findings suggest application of either low-risk rule would lead to decreased reliance on CT scan. Applying these clinical rules to their patients could have reduced the rate of CT scan by 20%. They conclude that pediatric patients who have suspected appendicitis and are at low risk by either model should be considered for observation rather than undergo CT scan or operative care.
Epidemiology of Fungal Infection in the Tertiary Care Inpatient Setting
Review by Brandan P. Kennedy, MD
Abelson, JA, Moore T, Bruckner D, et al. Frequency of fungemia in hospitalized pediatric inpatients over 11 years at a tertiary care institution. Pediatrics. 2005;116(1):611-617.
Fungal organisms are relatively uncommon causes of blood infections in the pediatric population. When they do occur, they cause significant morbidity and mortality. The incidence of fungal blood infections appears to be rising at a faster rate than that of other pathogens. Authors from the University of California performed this study to describe those rate changes, and to evaluate whether treatment has improved in the past 11 years.
This retrospective cohort study involved children admitted to Mattel Children’s Hospital in Los Angeles, California, during an 11-year period from January 1991 to December 2001. Information obtained on all positive fungal cultures from all body sites included date and site of culture, demographics, and fungal etiology. Additionally, data regarding underlying illness, hospital course, outcome, and antimicrobial treatment were considered.
Study results demonstrated a significant increase in diagnosed fungemia in children. There was a 15% increase in overall pediatric admissions in the study period, and a 23% increase in positive fungal cultures in the same period. Of 272 blood cultures, 97 were positive for fungus. Although this is a relatively small number of total infections, data demonstrated a 91% increase in fungemia during the study period. Candida species were the organisms most frequently isolated from any body site with 85% of the total isolates. Of the total positive isolates, it appeared that approximately 78% reflected colonization as opposed to infection.
Outcomes for fungal infections improved only marginally in the study period. Fifty percent of patients with fungemia died between 1991 and 1996, and 45% died between 1997 and 2001. The mortality rate for immunocompromised conditions was 57%. The cost effectiveness of fungal screening cultures was also evaluated, which demonstrated that fungal cultures identified 14 patients independent of bacterial blood cultures at a cost of $560,000, which resulted in a cost of $40,000 per identified patient.
The study reaches several significant conclusions. First there has been a significant increase in fungal infections that exceeds the increase in overall pediatric hospital admissions. However, it is important to recognize the increase in immunocompromised conditions during the study period, which may account for the higher incidence of fungemia.
Second the addition of fungal blood cultures to bacterial blood cultures as part of a routine workup for febrile patients appeared to yield limited clinical information at a very high cost. Third this study highlights the serious threat fungal infections pose to immunocompromised hosts who have a significantly higher incidence of infection, as well as higher morbidity and mortality. Fourth the use of broad spectrum antibiotics may be increasing fungal colonization in patients and consequently increasing the risk for pathologic fungal infection. Finally morbidity and mortality rates for fungal infections did not greatly improve, despite significant improvements in supportive care made during the study period
This study demonstrates a need for better diagnostic markers for fungal infections, especially those that might provide earlier detection and diagnosis at less cost. The importance of judicious use of antibiotics is underscored while the need for a broader base of therapeutic agents is highlighted. These issues may be key ingredients needed to reduce adverse outcomes from fungal infections, especially in the immunocompromised host.
Rotavirus Vaccine Revisited
Salinas B, Perez Schael I, Linhares AC, et al. Evaluation of safety, immunogenicity and efficacy of an attenuated rotavirus vaccine, RIX4414. Ped Infect Dis J. 2005;24(9):807-816.
Rotavirus is the leading cause of severe gastroenteritis among children worldwide. In the United States, rotavirus is responsible for approximately 5%-10% of all diarrhea among children older than five and accounts for approximately 50,000 hospitalizations each year. An estimated one in 200,000 children with rotavirus diarrhea dies from complications of infection. The immunizing effect of rotavirus infection stimulated the development of live attenuated vaccines. In 1998, a three-dose regimen of a tetravalent rhesus-human reassortant vaccine (RotaShield: Wyeth Laboratories,) was licensed for infant immunization in the United States. Within the first year of use, it was withdrawn due to an observed risk of intussusception. The current study was designed to evaluate immunogenicity and efficacy of a live attenuated monovalent human rotavirus vaccine, RIX4414.
A double-blind, randomized, placebo-controlled design was utilized with the RIX4414 rotavirus vaccine administered at three different virus concentrations. Infants were randomly assigned to one of the three study groups or the placebo group. Infants in the vaccine groups received two oral doses of the vaccine at the age of two and four months. An identical placebo containing the same constituents as the vaccine except for the vaccine virus was used as the control. The vaccine was given concomitantly with other routine vaccinations. To determine immunogenicity, blood samples were obtained from all infants immediately before the first vaccination to exclude previous rotavirus infection. Blood samples were obtained two months after the first second vaccine doses and again at one year to measure anti-rotavirus IgA antibodies. Additionally, stool was obtained from 25% of the study sample and tested for rotavirus viral shedding, with differentiation between wild type and vaccine also being performed.
There were 2,155 infants enrolled in this study from three countries in South America, allowing for slightly more than 500 infants in each group. The study began in May 2001, and the final one-year follow-up was completed in April 2003. The anti-rotavirus IgA seroconversion rates two months after first and second doses were 38%-43% and 61%-65%, respectively. This compared with a 5.3% seroconversion rate in the control group, which was determined to be a wild type virus. Vaccine take after two doses was shown in all three vaccine study groups, ranging from 65% to 75% for the lowest to highest vaccine concentration groups. Reactogenicity and safety was evaluated by monitoring incidences of fever, diarrhea, vomiting, irritability, and loss of appetite during the 15 days after vaccine administration. The results were similar for the four study groups suggesting no significant reactogenicity. There were 220 serious adverse events reported including one intussusception. However, none of these events was determined to be related to the vaccine. The RIX4414 vaccine was demonstrated to effectively protect against severe gastroenteritis caused by G1 type rotavirus and also was shown to provide some cross protection to other serotypes.
This well-designed study demonstrated a statistically significant reduction in gastroenteritis due to rotavirus infection, especially of the predominant G1 serotype, after two doses of the RIX4414 human rotavirus vaccine. Objective measures of viral shedding and IgA seroconversion support the efficacy of the vaccine. This study provided a large sample population with good controls.
An important and possibly confounding variable not addressed by the study was breastfeeding status of the infants. There is clinical evidence demonstrating the protective properties of secretory IgA in human milk against rotavirus infection, and this could have influenced the observed severity of disease in the sample population. It would have been interesting to isolate breastfeeding status as a study variable and note any effect on the results of the study.
Aside from this observation, this study appears to show a promising new oral rotavirus vaccine. As further research on the RIX4414 vaccine continues, there is hope that this vaccine could make a significant positive impact on morbidity and hospitalization rates for rotavirus infections worldwide. TH
Scenario: A 32-day-old female presents with vomiting and jaundice, and the mother reports that the child is irritable and not eating as well as usual. The pediatric hospitalist suspects a UTI and orders a dipstick/urinalysis and urine culture. The dipstick/urinalysis results are positive for nitrite and leukocyte esterase, indicating antibiotic therapy for presumed UTI.
Introduction
UTI is one of the most common bacterial infections in infants and young children. To prevent progression to pyelonephritis and avoid potential renal scarring or failure, early recognition and prompt treatment are critical.
Clinical signs and symptoms of UTI in newborns include jaundice, sepsis, failure to thrive, poor feeding, vomiting, and fever. In infants and preschoolers, hospitalists should also suspect UTI in the presence of diarrhea, strong-smelling urine, abdominal or flank pain, and new onset urinary incontinence.1
Treatment recommendations, which are age-dependent, include antibiotic therapy initiated upon an abnormal dipstick/urinalysis. If a urine culture is positive, a seven- to 14-day course of antibiotic therapy is recommended, followed by prophylactic antibiotics until results of imaging studies are available.1 According to a study by Hoberman and Wald, treatment of UTI with oral antibiotics alone is generally effective, even for young children with pyelonephritis.2
Imaging recommendations for a first UTI include ultrasound, cystogram, and renal cortical scan.
Up to this point, there’s been very little disagreement about the management of UTI. However, the question of whether the one-month-old patient in our scenario should be admitted or sent home with strict instructions on the administration of antibiotics remains controversial. The pediatric UTI guideline from Cincinnati Children’s Hospital Medical Center recommends routine hospital admission if the infant is under 30 days old.1 On the other hand, Santen and Altieri, among others, recommend, “Sick children and infants less than three months should be treated as inpatients, and healthy children and older infants may be treated as outpatients.”3,4
—Jeffrey M. Simmons, MD
The Set-Up
We asked several pediatric hospitalists across the country—including an instructor of pediatrics at Cincinnati Children’s Hospital Medical Center—to respond to this simple scenario, posing the following questions: “What recommendation would you follow? Would you admit the one-month-old infant in our scenario or send her home? Why?”
Predictably, admission criteria varied, but most agreed that this infant should be admitted. No one based their response on age.
Automatic Admission Cut-off Not Well Supported
Jeffrey M. Simmons, MD, instructor of pediatrics, Division of General and Community Pediatrics, Cincinnati Children’s Hospital Medical Center, responds: “My perspective on the UTI scenario is that the only dogma that applies is that the infant needs antibiotics. Because [the scenario] mentions vomiting, I would, therefore, most likely give parenteral antibiotics and admit.
“An issue within this scenario that I don’t believe the literature answers clearly is once a UTI is identified by the U/A, what risk remains of bacteremia or meningitis in the over 30-day-old infant? We are taught that infants don’t ‘localize’ infections well (i.e., a serious bacterial infection in one place can rapidly lead to disseminated infection). I sense a growing consensus that after 30 days or so this concern is less at issue. However, for clarity, I would prefer to obtain blood and CSF cultures on this infant prior to initiating antibiotic therapy. Without those cultures, and the screening tests that go with them (i.e., serum WBC count, CSF cell count, and glucose), I would be uncomfortable sending the infant home.
“However, if the WBC count was between 5 and 15, the CSF reassuring, the parents in agreement and reliable with good primary care follow-up the next morning, I think such an infant could be managed with one dose of IV/IM antibiotics after cultures are obtained and sent home. The following day, depending on culture results and the clinical situation, this infant might either be admitted, given another dose of parenteral antibiotics pending final blood culture, or switched to oral antibiotics.
“Such a plan is complex, so if the primary care physician or family was at all uncomfortable, admission to accomplish that plan would clearly be appropriate.
“A final issue not adequately addressed within the scenario is an assessment of the infant’s hydration—potentially an issue due to the vomiting. If the infant was mildly to moderately dehydrated with persistent vomiting, I would also then admit until this improved.
“In regard to the cited evidence, I would say that clearly an arbitrary cut-off of automatic admission under three months is not well supported by current literature, but is certainly many practitioners’ ‘style.’ Oral antibiotics are clearly efficacious for pyelonephritis, but the clinician needs to be confident the family can give them and the infant will keep them down. I believe most practitioners would agree with admission for an infant under 30 days, but where to precisely draw that line needs to be better established through investigation. Potential concomitant bacteremia and meningitis lead me to support obtaining blood and CSF cultures on any infant under 60 days for which I plan to initiate antibiotics for UTI.”
Possibility of Dehydration Indicates Admission
Michael P. Fullmer, DO, Central Iowa Pediatric Hospitalists, Mercy Medical Center, Des Moines, Iowa, responds: “This patient is not febrile, but does have other systemic symptoms, including poor feeding, irritability, jaundice, and vomiting. These systemic symptoms could be indicative of a serious bacterial infection like UTI, bacteremia, or meningitis. This patient most likely has a UTI. These symptoms may suggest pyelonephritis rather than lower UTI, but the distinction is not necessary for our decision here.
“Bacteremia is present in up to 22.7% of infants less than two months old with a UTI. This adds to the complexity of the issue, and a blood culture is probably indicated. Meningitis should be considered, but a lumbar puncture is probably not indicated in this scenario.
“A one-month-old with a UTI should be admitted to the hospital and started on parenteral antibiotics. There are several reasons for admission. First, the presence of vomiting makes oral antibiotic administration impractical (if not impossible). Intramuscular antibiotic injection may be an option, but the IV route gives the provider more options and is usually better accepted by parents. Next, if the patient is not already dehydrated, the poor feeding and vomiting could lead to dehydration. This alone would be an indication for admission. Finally, admission would give the physician time to observe the infant for clinical improvement. This may be subtle in the absence of fever.
“Another consideration is the recommended imaging for all children less than two years old. The AAP has recommended a renal ultrasound and VCUG for all infants and young children with their first UTI. This is important for discovery of urinary tract anomalies that predispose the patient to recurrent UTIs and eventual renal scarring and dysfunction. The imaging should be performed at the earliest convenient time if the patient is responding to therapy. In our practice, we generally have the renal ultrasound performed while the patient is in the hospital and arrange the VCUG as an outpatient [exam] prior to completion of the course of antibiotics.
“Patients are discharged when they are afebrile for 24 hours, have adequate oral intake and are able to take oral antibiotics. Please refer to the AAP Clinical Practice Guideline for more detail.5,6
—Michael Fullmer, DO
Admission Criteria Must Take Fever Into Account
John W. Graef, MD, chief, Services Office at Children’s Hospital, Harvard Vanguard Services Office, Boston, responds: “The presence or absence of fever is an important variable. Dipstick urines are shortcuts and don’t provide such information as the presence or absence of casts, although the fact that the child is vomiting and irritable is suggestive of pyelonephritis as opposed to a simple UTI/cystitis. [The scenario doesn’t] mention how the urine is obtained. Presumably it is a cath specimen, but that needs to be specified.
“Jaundice can occur with a UTI, but usually in the first week or so of life. A 32-day-old infant with pyelo is unlikely to be jaundiced unless for some other reason.
“I certainly agree with routine admission of a febrile infant up to one month, but not necessarily an otherwise well female infant. The presence of vomiting and irritability with or without fever might prompt a full septic work-up, in which case the decision to admit an afebrile infant would depend on the results of the CBC/UA and probably an LP. A blood culture and CBC should be drawn in the ED.
“In other words, one can’t have it both ways. If the irritability and vomiting are due to pyelo, a septic work-up is warranted. If all parameters are normal, the only reason for admission is hydration of a vomiting infant. If the infant was afebrile and had an uncomplicated UTI, I would not automatically admit an otherwise well 32-day-old.
“Poor PO (per OS, i.e., oral, by mouth) intake in an infant with a UTI warrants IV fluid regardless of age.”
—John W. Graef, MD
True Emesis With Decreased Oral Intake Indicates Admission
Erin R. Stucky, MD, pediatric hospitalist, Children’s Hospital and Health Center San Diego, associate professor, UCSD Pediatrics, responds: “The decision to admit or discharge this one-month-old with vomiting, jaundice, poor eating, and irritability should include consideration of hydration status, toxicity, sepsis potential, and ability to secure close follow-up. We will make the assumption that the infant is term, with no past medical history, family history, or prenatal evaluation that would put the infant at greater risk for sepsis or likelihood of underlying anatomic genitourinary abnormality. In addition we will assume that the urinalysis was performed by catheterization in a non-pretreated infant.
“A careful history should elicit the change in urine output, frequency, and volume of emesis to contrast with small ‘spit-ups,’ and change in feeding duration or volume. Observation of a feeding in the office or emergency department can be of great value in determining likelihood of ability to maintain hydration at home. Feeding type should be confirmed [because] parents may dilute or alternately concentrate formula in response to vomiting. In this scenario, true emesis in an infant with decreased oral intake would be grounds for admission, intravenous hydration, and evaluation of electrolytes.
“A newly irritable infant evokes a visceral sensation for all pediatric hospitalists. An irritable one-month-old may be suffering from a single system infection, electrolyte imbalance, or other insult, but unfortunately may just as easily have multisystem involvement. Vital signs and physical exam findings of toxicity, such as tachycardia and delayed capillary refill, are not known to us. The presence of jaundice raises concern for cholestasis induced by E. coli or possibly rarer metabolic disease, such as galactosemia. Admission would allow for evaluation and monitoring of the more likely causes of irritability in our index patient, which include sepsis, meningitis, and electrolyte and acid-base imbalances.
“The urinalysis in this infant is suggestive of a urinary tract infection, although infants may have no abnormalities noted on initial urinalysis.7,8 The risk of bacteremia in infants under 60 days with documented urinary tract infection is significant. A number of studies support the need to treat infants less than 30 days with parenteral antibiotics.9-14 Addition of C-reactive protein testing at this time does not aid in distinguishing those who are bacteremic in this age group.15 The often quoted study by Hoberman of 306 children included only 13 under the age of two months.16 Of the 13 reported positive blood cultures, 10 were in children under age six months. Daily intramuscular ceftriaxone treatment would cover a majority of the typical neonatal UTI organisms, can be administered in the outpatient setting, and is proven to be as effective as intravenous delivery. The clinical response to bacteremia is, however, unpredictable in young infants. The sepsis potential in this infant requires admission for physiologic monitoring and support as needed.
“Final, but not inconsequential, concerns are barriers to follow-up. These include parental experience and coping skills with feeding and monitoring an ill infant, ability to educate on the illness and reasons for follow-up, transportation, and operational issues, such as weekend clinic hours or holiday office closures. For the index patient these issues are overshadowed by the clinical criteria for admission but would be of great importance for discharge.”
Conclusion
Based on these responses, admitting a suspected UTI patient on the basis of age alone, as suggested by Santen and Altieri, is likely inappropriate. Many other factors must be weighed and would likely indicate admission for the patient in the scenario, regardless of the infant’s age. In short, until there’s better evidence for age-based admission criteria, clinical judgment based on the individual patient presentation must continue to drive care and treatment decisions.
Keri Losavio is a medical journalist with more than 10 years’ experience.
References
- UTI Guideline Team, Cincinnati Children’s Hospital Medical Center. “Evidence based clinical practice guideline for children 12 years of age or less with acute first time urinary tract infection.” www.cincinnatichildrens.org/svc/dept-div/health-policy/ev-based/uti.htm. Guideline 7, pages 1–20, April 2005.
- Hoberman A, Wald ER. Treatment of urinary tract infections. Pediatr Infect Dis J. 1999;18(11):1020–1021.
- Santen SA, Altieri MF. Pediatric urinary tract infection. Emerg Med Clin North Am. 200119(3):675–690.
- Egland AG, Egland TK. Pyelonephritis. eMedicine. www.emedicine.com/emerg/topic769.htm. Accessed Oct. 16, 2005.
- No authors listed. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Pediatrics. 1999;103(4):843–852.
- Pitetti RD, Choi S. Utility of blood cultures in febrile children with UTI. Am J Emerg Med. 2002;20:271–274.
- Dayan PS, Bennett J, Best R, et al. Test characteristics of the urine Gram stain in infants 60 days of age with fever. Pediatr Emerg Care. 2002;18(1):12–14.
- Huicho L, Campos-Sanchez M, Alamo C. Meta-analysis of urine screening tests for determining the risk of urinary tract infection in children. Pediatr Infect Dis J. 2002;21 (1):1-11.
- Byington C L, Enriquez F, Hoff C, et al. Serious bacterial infections in febrile infants 1 to 90 days old with and without viral infections. Pediatrics. 2004; 113(6):1662–1666.
- Baraff L. Management of fever without source in infants and children. Ann Emerg Med. 2000;36(6):602–614.
- Baraff LJ, Oslund SA, Schriger DL, Stephen ML. Probability of bacterial infections in febrile infants less than three months of age: A meta-analysis. Pediatr Infect Dis J. 1992;11(4):257–264.
- Klein JO. Management of the febrile child without a focus of infection in the era of universal pneumococcal immunization. Pediatr Infect Dis J. 2002;21(6):584–588.
- Syrogiannopoulos G, Grieva I, Anastassiou E, et al. Sterile cerebrospinal fluid pleocytosis in young infants with urinary tract infections. Pediatr Infect Dis J. 2001;20(10):927–930.
- Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febrile infants at low risk for serious bacterial infection—an appraisal of the Rochester criteria and implications for management. Febrile Collaborative Study Group. Pediatrics. 1994;94(3):390–396.
- Malik A, Hui C, Pennie RA, Kirpalani H. Beyond the complete blood cell count and C-reactive protein: A systematic review of modern diagnostic tests for neonatal sepsis. Arch Pediatr Adolesc Med. 2003;157(6):511–516.
- Hoberman A, Wald ER, Hickey RW, et al. Oral versus intravenous therapy for urinary tract Infections in young children. Pediatrics.1999;104:79–86
PEDIATRIC SPECIAL SECTION
IN THE LITERATURE
Utilize Clinical and Demographic Factors to Diagnose UTIs in Young Febrile Infants
Review by Sara E. Gardner, MD
Zorc JJ, Levine DA, Platt SL, et al. Clinical and demographic factors associated with urinary tract infections in young febrile infants. Pediatrics. 2000;116(3):644-648.
UTI is a common cause of serious bacterial infection in the febrile infant <60 days of age. Standard urinalysis and urine dipstick techniques, commonly used to diagnose UTI, have relatively low sensitivity increasing the possibility of a missed diagnosis. An accurate initial diagnosis is critical in this age group for whom complications from UTI include bacteremia and renal scarring.
To describe the demographic and clinical factors associated with UTI in infants ≤60 days of age with fever, these authors conducted a prospective cross sectional study from October 1999 to March 2001. Patients were enrolled at eight different institutions after presentation to an emergency department. One-thousand-twenty-five patients age 60 days or younger (mean age 35.5 days, 60.5% male) were enrolled with either reported or documented fever >38. Routine testing for all children included respiratory syncytial virus (RSV) sampling and bladder catheterization or suprapubic aspiration for urinalysis and culture.
A positive urinalysis was defined as a trace or greater result for leukocyte esterase and/or nitrite on dipstick or greater than or equal to five WBCs per high power field (hpf) on urine microscopy. UTI was defined as growth of a single pathogen of ≥1,000 colony forming units (cfu)/mL for urine cultures obtained by suprapubic catheterization, ≥50,000 cfu/mL from a catheterized specimen, or ≥10,000 cfu/ml from a catheterized specimen with a positive urinalysis.
Of the patients enrolled in the study, 92 were found to have UTI by these diagnostic criteria. Using the chi-squared test and calculated odds ratios with 95% confidence intervals, uncircumcised male (OR: 10.4; 95% CI: 4.7-31.4) and maximum temperature of ≥39º C (OR: 2.4 per degree C; 95% CI: 1.5-3.6) were found to be statistically significant variables for predicting UTI. These risk factors remained statistically significant after multivariable analysis controlling for other factors.
Interestingly of the above 92 patients diagnosed with UTI, 85 grew ≥50,000 cfu of a single pathogen, but six (8%) grew 10,000-49,000 cfu with a positive UA based on the study criteria. Zorc, et al. included these patients with >10,000 cfu and >5 WBC/hpf in this study despite previous studies that have established a definition of positive urinalysis to be ≥10 WBC/hpf. Zorc, et al. acknowledge the conservative definition applied in their current study, but assert that the overall results of the study would have been similar had 10,000 or 50,000 cfu/mL thresholds been chosen. To support this assertion, Zorc, et al. retrospectively applied enhanced urinalysis, a sensitive form of urinalysis including hemocytometric cell count and gram stain described by Hoberman, et al. to study patients with low bacteria counts. Based on Hoberman’s study, enhanced urinalysis can differentiate acute infection from asymptomatic bacteriuria in patients with bacterial growth between 10,000 to 50,000 cfu/mL.
Another significant limitation of this study was failure to enroll one-third of eligible patients to the study. In addition, the authors note that missed patients had a lower rate of UTI compared with enrolled patients.
Although this study design prohibits generalization to patient care areas outside the emergency department, the findings can assist the hospitalist in the evaluation of the febrile infant during RSV season and potentially guide decisions regarding empiric antibiotic therapy as part of evaluations to diagnose or exclude serious bacterial infection. Additionally, this study raises the question of need for better methods of urinalysis for febrile, uncircumcised male infants.
Back to the Basics: A Clinical Approach to Excluding Acute Appendicitis
Review by Jane G. Buss, MD
Kharbanda AB, Taylor GA, Fishman SJ, et al. A clinical decision rule to identify children at low risk for appendicitis. Pediatrics. 2005;116(3):709-16.
Appendicitis is the most common condition requiring emergency abdominal surgery in the pediatric population. To diagnosis appendicitis, clinicians typically utilize clinical findings, imaging studies, and laboratory testing. The use of clinical scoring systems to improve the diagnosis of appendicitis in children has been limited, mostly resulting from unacceptably low sensitivities/specificities, or the lack of validation. The use of CT scan to aid in the evaluation of children with appendicitis has become widespread. Concerns remain, however, regarding long-term radiation effects and increased healthcare costs associated with this approach.
The authors of this study sought to develop clinical scores for children to predict which of those with acute abdominal pain do not have appendicitis. Additionally, they hoped to lessen the use of CT scanning. Their goal was to identify those groups of children who have significant abdominal pain without appendicitis who could be safely observed without CT scan or possible surgery.
This prospective cohort study enrolled 601 eligible children ages three to 18 with suspected appendicitis who presented to the emergency department over a 15-month period. Two low-risk clinical decision rules were developed using logistic regression and recursive partitioning. Using logistic regression, six factors were identified from analysis of 425 patients in the derivation set significantly associated with an increased likelihood of appendicitis. The score components include:
- Nausea (2 points);
- History of focal right lower quadrant pain (RLQ) (2 points);
- Migration of pain (1 point);
- Difficulty walking (1 point);
- Rebound tenderness (2 points); and
- Absolute neutrophil count (ANC) >6.75 x 103/uL (6 points).
When tested with data from the 176 patients in the validation set, a score of less than or equal to five had a sensitivity of 96.3%, and a negative predictive value of 95.6% for excluding appendicitis.
The authors then derived a second clinical low-risk decision rule by recursive partitioning. They determined that a combination of ANC <6.75 X 10 to the third/microliter, absence of nausea (or emesis or anorexia), and absence of maximal tenderness in the RLQ essentially excluded appendicitis in the derivation and validation groups. This rule had a sensitivity of 98.1%, and a negative predictive value of 97.5%.
In summary, these authors derived two clinical decision rules giving the clinician the option of a clinical score (logistic regression) or a decision tree (recursive partitioning) to identify children at low risk for appendicitis. Their findings suggest application of either low-risk rule would lead to decreased reliance on CT scan. Applying these clinical rules to their patients could have reduced the rate of CT scan by 20%. They conclude that pediatric patients who have suspected appendicitis and are at low risk by either model should be considered for observation rather than undergo CT scan or operative care.
Epidemiology of Fungal Infection in the Tertiary Care Inpatient Setting
Review by Brandan P. Kennedy, MD
Abelson, JA, Moore T, Bruckner D, et al. Frequency of fungemia in hospitalized pediatric inpatients over 11 years at a tertiary care institution. Pediatrics. 2005;116(1):611-617.
Fungal organisms are relatively uncommon causes of blood infections in the pediatric population. When they do occur, they cause significant morbidity and mortality. The incidence of fungal blood infections appears to be rising at a faster rate than that of other pathogens. Authors from the University of California performed this study to describe those rate changes, and to evaluate whether treatment has improved in the past 11 years.
This retrospective cohort study involved children admitted to Mattel Children’s Hospital in Los Angeles, California, during an 11-year period from January 1991 to December 2001. Information obtained on all positive fungal cultures from all body sites included date and site of culture, demographics, and fungal etiology. Additionally, data regarding underlying illness, hospital course, outcome, and antimicrobial treatment were considered.
Study results demonstrated a significant increase in diagnosed fungemia in children. There was a 15% increase in overall pediatric admissions in the study period, and a 23% increase in positive fungal cultures in the same period. Of 272 blood cultures, 97 were positive for fungus. Although this is a relatively small number of total infections, data demonstrated a 91% increase in fungemia during the study period. Candida species were the organisms most frequently isolated from any body site with 85% of the total isolates. Of the total positive isolates, it appeared that approximately 78% reflected colonization as opposed to infection.
Outcomes for fungal infections improved only marginally in the study period. Fifty percent of patients with fungemia died between 1991 and 1996, and 45% died between 1997 and 2001. The mortality rate for immunocompromised conditions was 57%. The cost effectiveness of fungal screening cultures was also evaluated, which demonstrated that fungal cultures identified 14 patients independent of bacterial blood cultures at a cost of $560,000, which resulted in a cost of $40,000 per identified patient.
The study reaches several significant conclusions. First there has been a significant increase in fungal infections that exceeds the increase in overall pediatric hospital admissions. However, it is important to recognize the increase in immunocompromised conditions during the study period, which may account for the higher incidence of fungemia.
Second the addition of fungal blood cultures to bacterial blood cultures as part of a routine workup for febrile patients appeared to yield limited clinical information at a very high cost. Third this study highlights the serious threat fungal infections pose to immunocompromised hosts who have a significantly higher incidence of infection, as well as higher morbidity and mortality. Fourth the use of broad spectrum antibiotics may be increasing fungal colonization in patients and consequently increasing the risk for pathologic fungal infection. Finally morbidity and mortality rates for fungal infections did not greatly improve, despite significant improvements in supportive care made during the study period
This study demonstrates a need for better diagnostic markers for fungal infections, especially those that might provide earlier detection and diagnosis at less cost. The importance of judicious use of antibiotics is underscored while the need for a broader base of therapeutic agents is highlighted. These issues may be key ingredients needed to reduce adverse outcomes from fungal infections, especially in the immunocompromised host.
Rotavirus Vaccine Revisited
Salinas B, Perez Schael I, Linhares AC, et al. Evaluation of safety, immunogenicity and efficacy of an attenuated rotavirus vaccine, RIX4414. Ped Infect Dis J. 2005;24(9):807-816.
Rotavirus is the leading cause of severe gastroenteritis among children worldwide. In the United States, rotavirus is responsible for approximately 5%-10% of all diarrhea among children older than five and accounts for approximately 50,000 hospitalizations each year. An estimated one in 200,000 children with rotavirus diarrhea dies from complications of infection. The immunizing effect of rotavirus infection stimulated the development of live attenuated vaccines. In 1998, a three-dose regimen of a tetravalent rhesus-human reassortant vaccine (RotaShield: Wyeth Laboratories,) was licensed for infant immunization in the United States. Within the first year of use, it was withdrawn due to an observed risk of intussusception. The current study was designed to evaluate immunogenicity and efficacy of a live attenuated monovalent human rotavirus vaccine, RIX4414.
A double-blind, randomized, placebo-controlled design was utilized with the RIX4414 rotavirus vaccine administered at three different virus concentrations. Infants were randomly assigned to one of the three study groups or the placebo group. Infants in the vaccine groups received two oral doses of the vaccine at the age of two and four months. An identical placebo containing the same constituents as the vaccine except for the vaccine virus was used as the control. The vaccine was given concomitantly with other routine vaccinations. To determine immunogenicity, blood samples were obtained from all infants immediately before the first vaccination to exclude previous rotavirus infection. Blood samples were obtained two months after the first second vaccine doses and again at one year to measure anti-rotavirus IgA antibodies. Additionally, stool was obtained from 25% of the study sample and tested for rotavirus viral shedding, with differentiation between wild type and vaccine also being performed.
There were 2,155 infants enrolled in this study from three countries in South America, allowing for slightly more than 500 infants in each group. The study began in May 2001, and the final one-year follow-up was completed in April 2003. The anti-rotavirus IgA seroconversion rates two months after first and second doses were 38%-43% and 61%-65%, respectively. This compared with a 5.3% seroconversion rate in the control group, which was determined to be a wild type virus. Vaccine take after two doses was shown in all three vaccine study groups, ranging from 65% to 75% for the lowest to highest vaccine concentration groups. Reactogenicity and safety was evaluated by monitoring incidences of fever, diarrhea, vomiting, irritability, and loss of appetite during the 15 days after vaccine administration. The results were similar for the four study groups suggesting no significant reactogenicity. There were 220 serious adverse events reported including one intussusception. However, none of these events was determined to be related to the vaccine. The RIX4414 vaccine was demonstrated to effectively protect against severe gastroenteritis caused by G1 type rotavirus and also was shown to provide some cross protection to other serotypes.
This well-designed study demonstrated a statistically significant reduction in gastroenteritis due to rotavirus infection, especially of the predominant G1 serotype, after two doses of the RIX4414 human rotavirus vaccine. Objective measures of viral shedding and IgA seroconversion support the efficacy of the vaccine. This study provided a large sample population with good controls.
An important and possibly confounding variable not addressed by the study was breastfeeding status of the infants. There is clinical evidence demonstrating the protective properties of secretory IgA in human milk against rotavirus infection, and this could have influenced the observed severity of disease in the sample population. It would have been interesting to isolate breastfeeding status as a study variable and note any effect on the results of the study.
Aside from this observation, this study appears to show a promising new oral rotavirus vaccine. As further research on the RIX4414 vaccine continues, there is hope that this vaccine could make a significant positive impact on morbidity and hospitalization rates for rotavirus infections worldwide. TH
Scenario: A 32-day-old female presents with vomiting and jaundice, and the mother reports that the child is irritable and not eating as well as usual. The pediatric hospitalist suspects a UTI and orders a dipstick/urinalysis and urine culture. The dipstick/urinalysis results are positive for nitrite and leukocyte esterase, indicating antibiotic therapy for presumed UTI.
Introduction
UTI is one of the most common bacterial infections in infants and young children. To prevent progression to pyelonephritis and avoid potential renal scarring or failure, early recognition and prompt treatment are critical.
Clinical signs and symptoms of UTI in newborns include jaundice, sepsis, failure to thrive, poor feeding, vomiting, and fever. In infants and preschoolers, hospitalists should also suspect UTI in the presence of diarrhea, strong-smelling urine, abdominal or flank pain, and new onset urinary incontinence.1
Treatment recommendations, which are age-dependent, include antibiotic therapy initiated upon an abnormal dipstick/urinalysis. If a urine culture is positive, a seven- to 14-day course of antibiotic therapy is recommended, followed by prophylactic antibiotics until results of imaging studies are available.1 According to a study by Hoberman and Wald, treatment of UTI with oral antibiotics alone is generally effective, even for young children with pyelonephritis.2
Imaging recommendations for a first UTI include ultrasound, cystogram, and renal cortical scan.
Up to this point, there’s been very little disagreement about the management of UTI. However, the question of whether the one-month-old patient in our scenario should be admitted or sent home with strict instructions on the administration of antibiotics remains controversial. The pediatric UTI guideline from Cincinnati Children’s Hospital Medical Center recommends routine hospital admission if the infant is under 30 days old.1 On the other hand, Santen and Altieri, among others, recommend, “Sick children and infants less than three months should be treated as inpatients, and healthy children and older infants may be treated as outpatients.”3,4
—Jeffrey M. Simmons, MD
The Set-Up
We asked several pediatric hospitalists across the country—including an instructor of pediatrics at Cincinnati Children’s Hospital Medical Center—to respond to this simple scenario, posing the following questions: “What recommendation would you follow? Would you admit the one-month-old infant in our scenario or send her home? Why?”
Predictably, admission criteria varied, but most agreed that this infant should be admitted. No one based their response on age.
Automatic Admission Cut-off Not Well Supported
Jeffrey M. Simmons, MD, instructor of pediatrics, Division of General and Community Pediatrics, Cincinnati Children’s Hospital Medical Center, responds: “My perspective on the UTI scenario is that the only dogma that applies is that the infant needs antibiotics. Because [the scenario] mentions vomiting, I would, therefore, most likely give parenteral antibiotics and admit.
“An issue within this scenario that I don’t believe the literature answers clearly is once a UTI is identified by the U/A, what risk remains of bacteremia or meningitis in the over 30-day-old infant? We are taught that infants don’t ‘localize’ infections well (i.e., a serious bacterial infection in one place can rapidly lead to disseminated infection). I sense a growing consensus that after 30 days or so this concern is less at issue. However, for clarity, I would prefer to obtain blood and CSF cultures on this infant prior to initiating antibiotic therapy. Without those cultures, and the screening tests that go with them (i.e., serum WBC count, CSF cell count, and glucose), I would be uncomfortable sending the infant home.
“However, if the WBC count was between 5 and 15, the CSF reassuring, the parents in agreement and reliable with good primary care follow-up the next morning, I think such an infant could be managed with one dose of IV/IM antibiotics after cultures are obtained and sent home. The following day, depending on culture results and the clinical situation, this infant might either be admitted, given another dose of parenteral antibiotics pending final blood culture, or switched to oral antibiotics.
“Such a plan is complex, so if the primary care physician or family was at all uncomfortable, admission to accomplish that plan would clearly be appropriate.
“A final issue not adequately addressed within the scenario is an assessment of the infant’s hydration—potentially an issue due to the vomiting. If the infant was mildly to moderately dehydrated with persistent vomiting, I would also then admit until this improved.
“In regard to the cited evidence, I would say that clearly an arbitrary cut-off of automatic admission under three months is not well supported by current literature, but is certainly many practitioners’ ‘style.’ Oral antibiotics are clearly efficacious for pyelonephritis, but the clinician needs to be confident the family can give them and the infant will keep them down. I believe most practitioners would agree with admission for an infant under 30 days, but where to precisely draw that line needs to be better established through investigation. Potential concomitant bacteremia and meningitis lead me to support obtaining blood and CSF cultures on any infant under 60 days for which I plan to initiate antibiotics for UTI.”
Possibility of Dehydration Indicates Admission
Michael P. Fullmer, DO, Central Iowa Pediatric Hospitalists, Mercy Medical Center, Des Moines, Iowa, responds: “This patient is not febrile, but does have other systemic symptoms, including poor feeding, irritability, jaundice, and vomiting. These systemic symptoms could be indicative of a serious bacterial infection like UTI, bacteremia, or meningitis. This patient most likely has a UTI. These symptoms may suggest pyelonephritis rather than lower UTI, but the distinction is not necessary for our decision here.
“Bacteremia is present in up to 22.7% of infants less than two months old with a UTI. This adds to the complexity of the issue, and a blood culture is probably indicated. Meningitis should be considered, but a lumbar puncture is probably not indicated in this scenario.
“A one-month-old with a UTI should be admitted to the hospital and started on parenteral antibiotics. There are several reasons for admission. First, the presence of vomiting makes oral antibiotic administration impractical (if not impossible). Intramuscular antibiotic injection may be an option, but the IV route gives the provider more options and is usually better accepted by parents. Next, if the patient is not already dehydrated, the poor feeding and vomiting could lead to dehydration. This alone would be an indication for admission. Finally, admission would give the physician time to observe the infant for clinical improvement. This may be subtle in the absence of fever.
“Another consideration is the recommended imaging for all children less than two years old. The AAP has recommended a renal ultrasound and VCUG for all infants and young children with their first UTI. This is important for discovery of urinary tract anomalies that predispose the patient to recurrent UTIs and eventual renal scarring and dysfunction. The imaging should be performed at the earliest convenient time if the patient is responding to therapy. In our practice, we generally have the renal ultrasound performed while the patient is in the hospital and arrange the VCUG as an outpatient [exam] prior to completion of the course of antibiotics.
“Patients are discharged when they are afebrile for 24 hours, have adequate oral intake and are able to take oral antibiotics. Please refer to the AAP Clinical Practice Guideline for more detail.5,6
—Michael Fullmer, DO
Admission Criteria Must Take Fever Into Account
John W. Graef, MD, chief, Services Office at Children’s Hospital, Harvard Vanguard Services Office, Boston, responds: “The presence or absence of fever is an important variable. Dipstick urines are shortcuts and don’t provide such information as the presence or absence of casts, although the fact that the child is vomiting and irritable is suggestive of pyelonephritis as opposed to a simple UTI/cystitis. [The scenario doesn’t] mention how the urine is obtained. Presumably it is a cath specimen, but that needs to be specified.
“Jaundice can occur with a UTI, but usually in the first week or so of life. A 32-day-old infant with pyelo is unlikely to be jaundiced unless for some other reason.
“I certainly agree with routine admission of a febrile infant up to one month, but not necessarily an otherwise well female infant. The presence of vomiting and irritability with or without fever might prompt a full septic work-up, in which case the decision to admit an afebrile infant would depend on the results of the CBC/UA and probably an LP. A blood culture and CBC should be drawn in the ED.
“In other words, one can’t have it both ways. If the irritability and vomiting are due to pyelo, a septic work-up is warranted. If all parameters are normal, the only reason for admission is hydration of a vomiting infant. If the infant was afebrile and had an uncomplicated UTI, I would not automatically admit an otherwise well 32-day-old.
“Poor PO (per OS, i.e., oral, by mouth) intake in an infant with a UTI warrants IV fluid regardless of age.”
—John W. Graef, MD
True Emesis With Decreased Oral Intake Indicates Admission
Erin R. Stucky, MD, pediatric hospitalist, Children’s Hospital and Health Center San Diego, associate professor, UCSD Pediatrics, responds: “The decision to admit or discharge this one-month-old with vomiting, jaundice, poor eating, and irritability should include consideration of hydration status, toxicity, sepsis potential, and ability to secure close follow-up. We will make the assumption that the infant is term, with no past medical history, family history, or prenatal evaluation that would put the infant at greater risk for sepsis or likelihood of underlying anatomic genitourinary abnormality. In addition we will assume that the urinalysis was performed by catheterization in a non-pretreated infant.
“A careful history should elicit the change in urine output, frequency, and volume of emesis to contrast with small ‘spit-ups,’ and change in feeding duration or volume. Observation of a feeding in the office or emergency department can be of great value in determining likelihood of ability to maintain hydration at home. Feeding type should be confirmed [because] parents may dilute or alternately concentrate formula in response to vomiting. In this scenario, true emesis in an infant with decreased oral intake would be grounds for admission, intravenous hydration, and evaluation of electrolytes.
“A newly irritable infant evokes a visceral sensation for all pediatric hospitalists. An irritable one-month-old may be suffering from a single system infection, electrolyte imbalance, or other insult, but unfortunately may just as easily have multisystem involvement. Vital signs and physical exam findings of toxicity, such as tachycardia and delayed capillary refill, are not known to us. The presence of jaundice raises concern for cholestasis induced by E. coli or possibly rarer metabolic disease, such as galactosemia. Admission would allow for evaluation and monitoring of the more likely causes of irritability in our index patient, which include sepsis, meningitis, and electrolyte and acid-base imbalances.
“The urinalysis in this infant is suggestive of a urinary tract infection, although infants may have no abnormalities noted on initial urinalysis.7,8 The risk of bacteremia in infants under 60 days with documented urinary tract infection is significant. A number of studies support the need to treat infants less than 30 days with parenteral antibiotics.9-14 Addition of C-reactive protein testing at this time does not aid in distinguishing those who are bacteremic in this age group.15 The often quoted study by Hoberman of 306 children included only 13 under the age of two months.16 Of the 13 reported positive blood cultures, 10 were in children under age six months. Daily intramuscular ceftriaxone treatment would cover a majority of the typical neonatal UTI organisms, can be administered in the outpatient setting, and is proven to be as effective as intravenous delivery. The clinical response to bacteremia is, however, unpredictable in young infants. The sepsis potential in this infant requires admission for physiologic monitoring and support as needed.
“Final, but not inconsequential, concerns are barriers to follow-up. These include parental experience and coping skills with feeding and monitoring an ill infant, ability to educate on the illness and reasons for follow-up, transportation, and operational issues, such as weekend clinic hours or holiday office closures. For the index patient these issues are overshadowed by the clinical criteria for admission but would be of great importance for discharge.”
Conclusion
Based on these responses, admitting a suspected UTI patient on the basis of age alone, as suggested by Santen and Altieri, is likely inappropriate. Many other factors must be weighed and would likely indicate admission for the patient in the scenario, regardless of the infant’s age. In short, until there’s better evidence for age-based admission criteria, clinical judgment based on the individual patient presentation must continue to drive care and treatment decisions.
Keri Losavio is a medical journalist with more than 10 years’ experience.
References
- UTI Guideline Team, Cincinnati Children’s Hospital Medical Center. “Evidence based clinical practice guideline for children 12 years of age or less with acute first time urinary tract infection.” www.cincinnatichildrens.org/svc/dept-div/health-policy/ev-based/uti.htm. Guideline 7, pages 1–20, April 2005.
- Hoberman A, Wald ER. Treatment of urinary tract infections. Pediatr Infect Dis J. 1999;18(11):1020–1021.
- Santen SA, Altieri MF. Pediatric urinary tract infection. Emerg Med Clin North Am. 200119(3):675–690.
- Egland AG, Egland TK. Pyelonephritis. eMedicine. www.emedicine.com/emerg/topic769.htm. Accessed Oct. 16, 2005.
- No authors listed. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Pediatrics. 1999;103(4):843–852.
- Pitetti RD, Choi S. Utility of blood cultures in febrile children with UTI. Am J Emerg Med. 2002;20:271–274.
- Dayan PS, Bennett J, Best R, et al. Test characteristics of the urine Gram stain in infants 60 days of age with fever. Pediatr Emerg Care. 2002;18(1):12–14.
- Huicho L, Campos-Sanchez M, Alamo C. Meta-analysis of urine screening tests for determining the risk of urinary tract infection in children. Pediatr Infect Dis J. 2002;21 (1):1-11.
- Byington C L, Enriquez F, Hoff C, et al. Serious bacterial infections in febrile infants 1 to 90 days old with and without viral infections. Pediatrics. 2004; 113(6):1662–1666.
- Baraff L. Management of fever without source in infants and children. Ann Emerg Med. 2000;36(6):602–614.
- Baraff LJ, Oslund SA, Schriger DL, Stephen ML. Probability of bacterial infections in febrile infants less than three months of age: A meta-analysis. Pediatr Infect Dis J. 1992;11(4):257–264.
- Klein JO. Management of the febrile child without a focus of infection in the era of universal pneumococcal immunization. Pediatr Infect Dis J. 2002;21(6):584–588.
- Syrogiannopoulos G, Grieva I, Anastassiou E, et al. Sterile cerebrospinal fluid pleocytosis in young infants with urinary tract infections. Pediatr Infect Dis J. 2001;20(10):927–930.
- Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febrile infants at low risk for serious bacterial infection—an appraisal of the Rochester criteria and implications for management. Febrile Collaborative Study Group. Pediatrics. 1994;94(3):390–396.
- Malik A, Hui C, Pennie RA, Kirpalani H. Beyond the complete blood cell count and C-reactive protein: A systematic review of modern diagnostic tests for neonatal sepsis. Arch Pediatr Adolesc Med. 2003;157(6):511–516.
- Hoberman A, Wald ER, Hickey RW, et al. Oral versus intravenous therapy for urinary tract Infections in young children. Pediatrics.1999;104:79–86
PEDIATRIC SPECIAL SECTION
IN THE LITERATURE
Utilize Clinical and Demographic Factors to Diagnose UTIs in Young Febrile Infants
Review by Sara E. Gardner, MD
Zorc JJ, Levine DA, Platt SL, et al. Clinical and demographic factors associated with urinary tract infections in young febrile infants. Pediatrics. 2000;116(3):644-648.
UTI is a common cause of serious bacterial infection in the febrile infant <60 days of age. Standard urinalysis and urine dipstick techniques, commonly used to diagnose UTI, have relatively low sensitivity increasing the possibility of a missed diagnosis. An accurate initial diagnosis is critical in this age group for whom complications from UTI include bacteremia and renal scarring.
To describe the demographic and clinical factors associated with UTI in infants ≤60 days of age with fever, these authors conducted a prospective cross sectional study from October 1999 to March 2001. Patients were enrolled at eight different institutions after presentation to an emergency department. One-thousand-twenty-five patients age 60 days or younger (mean age 35.5 days, 60.5% male) were enrolled with either reported or documented fever >38. Routine testing for all children included respiratory syncytial virus (RSV) sampling and bladder catheterization or suprapubic aspiration for urinalysis and culture.
A positive urinalysis was defined as a trace or greater result for leukocyte esterase and/or nitrite on dipstick or greater than or equal to five WBCs per high power field (hpf) on urine microscopy. UTI was defined as growth of a single pathogen of ≥1,000 colony forming units (cfu)/mL for urine cultures obtained by suprapubic catheterization, ≥50,000 cfu/mL from a catheterized specimen, or ≥10,000 cfu/ml from a catheterized specimen with a positive urinalysis.
Of the patients enrolled in the study, 92 were found to have UTI by these diagnostic criteria. Using the chi-squared test and calculated odds ratios with 95% confidence intervals, uncircumcised male (OR: 10.4; 95% CI: 4.7-31.4) and maximum temperature of ≥39º C (OR: 2.4 per degree C; 95% CI: 1.5-3.6) were found to be statistically significant variables for predicting UTI. These risk factors remained statistically significant after multivariable analysis controlling for other factors.
Interestingly of the above 92 patients diagnosed with UTI, 85 grew ≥50,000 cfu of a single pathogen, but six (8%) grew 10,000-49,000 cfu with a positive UA based on the study criteria. Zorc, et al. included these patients with >10,000 cfu and >5 WBC/hpf in this study despite previous studies that have established a definition of positive urinalysis to be ≥10 WBC/hpf. Zorc, et al. acknowledge the conservative definition applied in their current study, but assert that the overall results of the study would have been similar had 10,000 or 50,000 cfu/mL thresholds been chosen. To support this assertion, Zorc, et al. retrospectively applied enhanced urinalysis, a sensitive form of urinalysis including hemocytometric cell count and gram stain described by Hoberman, et al. to study patients with low bacteria counts. Based on Hoberman’s study, enhanced urinalysis can differentiate acute infection from asymptomatic bacteriuria in patients with bacterial growth between 10,000 to 50,000 cfu/mL.
Another significant limitation of this study was failure to enroll one-third of eligible patients to the study. In addition, the authors note that missed patients had a lower rate of UTI compared with enrolled patients.
Although this study design prohibits generalization to patient care areas outside the emergency department, the findings can assist the hospitalist in the evaluation of the febrile infant during RSV season and potentially guide decisions regarding empiric antibiotic therapy as part of evaluations to diagnose or exclude serious bacterial infection. Additionally, this study raises the question of need for better methods of urinalysis for febrile, uncircumcised male infants.
Back to the Basics: A Clinical Approach to Excluding Acute Appendicitis
Review by Jane G. Buss, MD
Kharbanda AB, Taylor GA, Fishman SJ, et al. A clinical decision rule to identify children at low risk for appendicitis. Pediatrics. 2005;116(3):709-16.
Appendicitis is the most common condition requiring emergency abdominal surgery in the pediatric population. To diagnosis appendicitis, clinicians typically utilize clinical findings, imaging studies, and laboratory testing. The use of clinical scoring systems to improve the diagnosis of appendicitis in children has been limited, mostly resulting from unacceptably low sensitivities/specificities, or the lack of validation. The use of CT scan to aid in the evaluation of children with appendicitis has become widespread. Concerns remain, however, regarding long-term radiation effects and increased healthcare costs associated with this approach.
The authors of this study sought to develop clinical scores for children to predict which of those with acute abdominal pain do not have appendicitis. Additionally, they hoped to lessen the use of CT scanning. Their goal was to identify those groups of children who have significant abdominal pain without appendicitis who could be safely observed without CT scan or possible surgery.
This prospective cohort study enrolled 601 eligible children ages three to 18 with suspected appendicitis who presented to the emergency department over a 15-month period. Two low-risk clinical decision rules were developed using logistic regression and recursive partitioning. Using logistic regression, six factors were identified from analysis of 425 patients in the derivation set significantly associated with an increased likelihood of appendicitis. The score components include:
- Nausea (2 points);
- History of focal right lower quadrant pain (RLQ) (2 points);
- Migration of pain (1 point);
- Difficulty walking (1 point);
- Rebound tenderness (2 points); and
- Absolute neutrophil count (ANC) >6.75 x 103/uL (6 points).
When tested with data from the 176 patients in the validation set, a score of less than or equal to five had a sensitivity of 96.3%, and a negative predictive value of 95.6% for excluding appendicitis.
The authors then derived a second clinical low-risk decision rule by recursive partitioning. They determined that a combination of ANC <6.75 X 10 to the third/microliter, absence of nausea (or emesis or anorexia), and absence of maximal tenderness in the RLQ essentially excluded appendicitis in the derivation and validation groups. This rule had a sensitivity of 98.1%, and a negative predictive value of 97.5%.
In summary, these authors derived two clinical decision rules giving the clinician the option of a clinical score (logistic regression) or a decision tree (recursive partitioning) to identify children at low risk for appendicitis. Their findings suggest application of either low-risk rule would lead to decreased reliance on CT scan. Applying these clinical rules to their patients could have reduced the rate of CT scan by 20%. They conclude that pediatric patients who have suspected appendicitis and are at low risk by either model should be considered for observation rather than undergo CT scan or operative care.
Epidemiology of Fungal Infection in the Tertiary Care Inpatient Setting
Review by Brandan P. Kennedy, MD
Abelson, JA, Moore T, Bruckner D, et al. Frequency of fungemia in hospitalized pediatric inpatients over 11 years at a tertiary care institution. Pediatrics. 2005;116(1):611-617.
Fungal organisms are relatively uncommon causes of blood infections in the pediatric population. When they do occur, they cause significant morbidity and mortality. The incidence of fungal blood infections appears to be rising at a faster rate than that of other pathogens. Authors from the University of California performed this study to describe those rate changes, and to evaluate whether treatment has improved in the past 11 years.
This retrospective cohort study involved children admitted to Mattel Children’s Hospital in Los Angeles, California, during an 11-year period from January 1991 to December 2001. Information obtained on all positive fungal cultures from all body sites included date and site of culture, demographics, and fungal etiology. Additionally, data regarding underlying illness, hospital course, outcome, and antimicrobial treatment were considered.
Study results demonstrated a significant increase in diagnosed fungemia in children. There was a 15% increase in overall pediatric admissions in the study period, and a 23% increase in positive fungal cultures in the same period. Of 272 blood cultures, 97 were positive for fungus. Although this is a relatively small number of total infections, data demonstrated a 91% increase in fungemia during the study period. Candida species were the organisms most frequently isolated from any body site with 85% of the total isolates. Of the total positive isolates, it appeared that approximately 78% reflected colonization as opposed to infection.
Outcomes for fungal infections improved only marginally in the study period. Fifty percent of patients with fungemia died between 1991 and 1996, and 45% died between 1997 and 2001. The mortality rate for immunocompromised conditions was 57%. The cost effectiveness of fungal screening cultures was also evaluated, which demonstrated that fungal cultures identified 14 patients independent of bacterial blood cultures at a cost of $560,000, which resulted in a cost of $40,000 per identified patient.
The study reaches several significant conclusions. First there has been a significant increase in fungal infections that exceeds the increase in overall pediatric hospital admissions. However, it is important to recognize the increase in immunocompromised conditions during the study period, which may account for the higher incidence of fungemia.
Second the addition of fungal blood cultures to bacterial blood cultures as part of a routine workup for febrile patients appeared to yield limited clinical information at a very high cost. Third this study highlights the serious threat fungal infections pose to immunocompromised hosts who have a significantly higher incidence of infection, as well as higher morbidity and mortality. Fourth the use of broad spectrum antibiotics may be increasing fungal colonization in patients and consequently increasing the risk for pathologic fungal infection. Finally morbidity and mortality rates for fungal infections did not greatly improve, despite significant improvements in supportive care made during the study period
This study demonstrates a need for better diagnostic markers for fungal infections, especially those that might provide earlier detection and diagnosis at less cost. The importance of judicious use of antibiotics is underscored while the need for a broader base of therapeutic agents is highlighted. These issues may be key ingredients needed to reduce adverse outcomes from fungal infections, especially in the immunocompromised host.
Rotavirus Vaccine Revisited
Salinas B, Perez Schael I, Linhares AC, et al. Evaluation of safety, immunogenicity and efficacy of an attenuated rotavirus vaccine, RIX4414. Ped Infect Dis J. 2005;24(9):807-816.
Rotavirus is the leading cause of severe gastroenteritis among children worldwide. In the United States, rotavirus is responsible for approximately 5%-10% of all diarrhea among children older than five and accounts for approximately 50,000 hospitalizations each year. An estimated one in 200,000 children with rotavirus diarrhea dies from complications of infection. The immunizing effect of rotavirus infection stimulated the development of live attenuated vaccines. In 1998, a three-dose regimen of a tetravalent rhesus-human reassortant vaccine (RotaShield: Wyeth Laboratories,) was licensed for infant immunization in the United States. Within the first year of use, it was withdrawn due to an observed risk of intussusception. The current study was designed to evaluate immunogenicity and efficacy of a live attenuated monovalent human rotavirus vaccine, RIX4414.
A double-blind, randomized, placebo-controlled design was utilized with the RIX4414 rotavirus vaccine administered at three different virus concentrations. Infants were randomly assigned to one of the three study groups or the placebo group. Infants in the vaccine groups received two oral doses of the vaccine at the age of two and four months. An identical placebo containing the same constituents as the vaccine except for the vaccine virus was used as the control. The vaccine was given concomitantly with other routine vaccinations. To determine immunogenicity, blood samples were obtained from all infants immediately before the first vaccination to exclude previous rotavirus infection. Blood samples were obtained two months after the first second vaccine doses and again at one year to measure anti-rotavirus IgA antibodies. Additionally, stool was obtained from 25% of the study sample and tested for rotavirus viral shedding, with differentiation between wild type and vaccine also being performed.
There were 2,155 infants enrolled in this study from three countries in South America, allowing for slightly more than 500 infants in each group. The study began in May 2001, and the final one-year follow-up was completed in April 2003. The anti-rotavirus IgA seroconversion rates two months after first and second doses were 38%-43% and 61%-65%, respectively. This compared with a 5.3% seroconversion rate in the control group, which was determined to be a wild type virus. Vaccine take after two doses was shown in all three vaccine study groups, ranging from 65% to 75% for the lowest to highest vaccine concentration groups. Reactogenicity and safety was evaluated by monitoring incidences of fever, diarrhea, vomiting, irritability, and loss of appetite during the 15 days after vaccine administration. The results were similar for the four study groups suggesting no significant reactogenicity. There were 220 serious adverse events reported including one intussusception. However, none of these events was determined to be related to the vaccine. The RIX4414 vaccine was demonstrated to effectively protect against severe gastroenteritis caused by G1 type rotavirus and also was shown to provide some cross protection to other serotypes.
This well-designed study demonstrated a statistically significant reduction in gastroenteritis due to rotavirus infection, especially of the predominant G1 serotype, after two doses of the RIX4414 human rotavirus vaccine. Objective measures of viral shedding and IgA seroconversion support the efficacy of the vaccine. This study provided a large sample population with good controls.
An important and possibly confounding variable not addressed by the study was breastfeeding status of the infants. There is clinical evidence demonstrating the protective properties of secretory IgA in human milk against rotavirus infection, and this could have influenced the observed severity of disease in the sample population. It would have been interesting to isolate breastfeeding status as a study variable and note any effect on the results of the study.
Aside from this observation, this study appears to show a promising new oral rotavirus vaccine. As further research on the RIX4414 vaccine continues, there is hope that this vaccine could make a significant positive impact on morbidity and hospitalization rates for rotavirus infections worldwide. TH