Challenging Dogma in the Treatment of Childhood Infections: Oral Antibiotics and Shorter Durations

Intravenous (IV) antibiotic therapy is a mainstay of treatment for numerous bacterial infections, but is associated with longer duration of inpatient hospitalization, greater healthcare costs, opportunities for line infection and other complications, and potentially reduced patient quality-of-life, compared with oral antibiotic therapy.1 In children, establishing and maintaining IV access can be particularly challenging and potentially traumatic. Transition to oral antibiotic therapy is therefore important but unwarranted delays occur frequently. Prescribed durations of antibiotic therapy, for both adults and children, are often based on historical dogma rather than high-quality evidence. However, a substantial body of clinical trials has now established that oral therapy is at least as effective as IV-only therapy in adults with various types of infections, and limited research has also been conducted for the treatment of common neonatal and pediatric diseases, alongside shorter durations of total therapy. The Australian and New Zealand Paediatric Infectious Diseases group published a systematic review and guidelines on this topic in 2016,2 while similar efforts are underway in adult medicine.3 In this review we consider recent developments in this area, with studies that challenge longstanding practices with respect to IV-to-oral antibiotic switch and the duration of therapy in immunocompetent children and neonates. Urinary Tract Infections Urinary tract infection (UTI) is common in children, with IV antibiotics generally recommended for infants aged 3 days of IV antibiotics.4 Previous studies, largely observational, support entirely oral therapy or rapid oral switch for children with UTI, even those with pyelonephritis.2 In contrast, a recent randomized controlled trial (RCT) investigated the efficacy of "standard"-course (10 days) and short-course (5 days) therapy for children in the United States aged 2 months to 10 years with UTI.5 Among 664 patients randomized, the largest such trial to date, children receiving short-course therapy were significantly more likely to have symptomatic UTI or asymptomatic bacteriuria between 6 days and the 11–14 day follow-up visit, after starting treatment. The outcome of treatment failure occurred infrequently in the short-course group (4.2%), however for patients receiving standard care, measurement of the primary outcome took place immediately after cessation of therapy. The authors acknowledge that this likely introduced bias favoring the standard course. Furthermore, there was no difference in treatment failure after the placebo versus antibiotic treatment period ended. The authors therefore concluded that short-course therapy could be "considered as a reasonable option for children exhibiting clinical improvement after 5 days of antimicrobial treatment." More recently, a smaller RCT was published comparing the management of febrile UTI in 142 well-appearing children with either 5 or 10 days of amoxicillin-clavulanate.6 The primary outcome, rate of recurrence, was significantly lower in the 5-day group (14.3% vs. 2.8%), strengthening the evidence for shorter duration therapy in clinically stable children. Although clinicians may be cautious about oral therapy for children with bacteremia or anatomical variations of the genitourinary tract, including vesicoureteral reflux, lack of evidence to support longer or all-IV therapy is notable and should not dissuade consideration of IV-to-oral antibiotic switch in future research and guidelines. Bone and Joint Infections The management of osteomyelitis and septic arthritis in patients of all ages has conventionally included the prolonged administration of IV antibiotics, with adults often treated with at least 6 weeks of IV therapy. Historically, children have received similar treatment, although shorter IV courses with oral switch and reduced duration of total therapy are now common: as short as 3 days of IV antibiotics, followed by oral therapy.2 In children who present with bone and joint infections in the absence of sepsis, it is possible that even shorter or entirely oral antibiotic therapy may be noninferior to standard therapy, and clinical trials of entirely oral therapy for bone and joint infection in children are currently underway.7 Pediatric Bacteremia and Meningitis For bloodstream infections in children, entirely IV therapy is generally recommended, however, this does not typically include children with bloodstream infections in the context of bone and joint infections, UTIs and pneumococcal infections, where early oral switch may be recommended.2 Again, durations and routes of therapy are often based on conventional wisdom, rather than high-quality evidence. Reliance on IV therapy for bacteremia and even endocarditis has also been challenged, with trials and large retrospective studies available to support similar efficacy outcomes in selected patients, with shorter hospitalization and fewer adverse events.8 It should be noted that this evidence is emerging from studies undertaken in adults, with limited pediatric data published to date. Traditional durations of therapy for meningitis in children have also been challenged, with open-label trials published over many years, predominantly in resource-constrained settings, using therapy as short as 4 days in selected patients and finding comparable outcomes for courses of 4 to 7 days and >7 to 14 days.2 These studies have not resulted in widespread change in practice, however, potentially due to concerns about selection bias, including low mortality in included patients relative to expected outcomes, the absence of identified causative organisms, or a high prevalence of Haemophilus influenzae B meningitis, which has precipitously declined in areas where infant Haemophilus influenzae B immunization is available. Further trials of therapy for meningitis should be undertaken, accounting for causative organisms.2 Appendicitis Appendicitis is the most common pediatric surgical diagnosis. While uncomplicated appendicitis can be managed operatively with single-dose surgical prophylaxis, nonoperative antibiotic management of early appendicitis is increasingly common. Current trends and controversies in the nonoperative management of appendicitis in children have been recently reviewed in detail, but did not include detail on antibiotic route or duration.9 One recent cohort study found that, in 67.1% of children, initial nonoperative management with a minimum 24 hours of IV antibiotics and 7 days of total antibiotic therapy effectively treated appendicitis without the need for subsequent appendectomy, with statistically significantly fewer disability days at 1 year relative to urgent surgical management.10 Not all cases of uncomplicated appendicitis are suitable for nonoperative management, however, with risk of perforation and treatment failure and a lower overall cure rate important considerations.11 Management of complicated appendicitis, including patients with perforated appendix and peritonitis, generally includes several days of initial IV therapy in the hospital. Australian guidelines recommend IV-to-oral antibiotic switch once the patient is afebrile and has normal bowel function.2 Recently published data regarding complicated appendicitis in both adults and children aged ≥8 years has demonstrated that 2-day courses of IV therapy are noninferior to 5 days of IV therapy in patients undergoing laparoscopic appendectomy, with similar rates of reintervention and complication.12 Community-acquired Pneumonia Community-acquired pneumonia (CAP) is a leading cause of mortality globally in children. The causative organism in CAP is often difficult to identify, particularly in children, with antibiotics generally given empirically, leading to uncertainty about treatment efficacy. While children with CAP are often treated with entirely oral therapy already, guidance on antibiotic duration has been developed with limited evidence. Recent trials have attempted to address the evidence gap. A 2023 systematic review and meta-analysis of 16 trials and 12,774 patients in total, from low and middle-income countries and high-income countries, found no substantial differences between short and longer-duration antibiotic therapy in terms of clinical cure, treatment failure or serious adverse events.13 Notable recent trials for CAP in children are listed in Table 1. A caveat is that all these studies included children with mild-moderate CAP, and optimal duration and timing of oral switch in patients with severe CAP remains an area requiring further research. Children may also have had viral pneumonia, which may have biased results in favor of noninferiority, and better tests to distinguish bacterial from viral illnesses in children are needed. TABLE 1. - Summary of Recent Studies Evaluating Intravenous-to-oral Antibiotic Switch and Shorter Oral Therapy for Children and Neonates Reference Diagnosis Study Type Setting (Country, Level of Care) Measure Patient Population No. of Patients Results Studies investigating duration of antibiotic therapy (including efficacy in nonoperative management) 4 UTI (bacteremic) Systematic review Multiple, multilevel ≤7 vs. >7 days of IV therapy Infants aged ≤90 days with UTI 6 studies Total 468 patients 2 studies describing 30-day UTI recurrence found no significant association with IV duration 2 studies describing 30-day all-cause hospitalization found no significant association with IV duration 4 UTI (nonbacteremic) Systematic review Multiple, multilevel ≤3 vs. >3 days of IV therapy Infants aged ≤90 days with UTI 12 studies Total 15,826 patients 2 studies describing 30-day UTI recurrence found no significant association with IV duration 4 studies describing UTI complications found no association with IV duration 5 UTI RCT United States, tertiary 5 vs. 10 days of oral therapy Children aged 2 months—10 years with UTI showing clinical improvement by day 5 of antibiotics 664 5 days oral therapy was associated with a greater risk of UTI on days 6–14 (4.2% vs. 0.6%, 1-sided 95% CI: 5.5; P 5 days of oral therapy Children with CAP treated as outpatients with oral antibiotics 16 trials Total 12,774 patients No difference in rates of clinical cure (OR: 1.01; 95% CI: 0.87–1.17) No difference in rates of relapse (RR: 1.12; 95% CI: 0.92–1.35) No difference in risk of hospitalization (RD: −0.2%; 95% CI: −0.9–0.5) Studies investigating IV-to-oral antibiotic switch 17 Neonatal infection Systematic review and meta-analysis Multiple, multilevel Oral antibiotic switch vs. IV therapy Neonates treated with oral antibiotics 31 studies total Total number of patients not described by authors No significant difference in relapse rates (OR: 0.95; 95% CI: 0.79–1.16; I 2: 0%) No significant difference in mortality (OR: 1.11; 95% CI: 0.72–1.72; I 2: 0%) 18 Neonatal infection (probable) RCT The Netherlands, tertiary Oral switch vs. 7 days of IV therapy Term neonates with probable bacterial infection 510 No significant difference in rate of reinfection (Absolute difference 0%; 95% CI: −1.9–1.9; P < 0.0001) Shorter median hospital length-of-stay in oral switch group. (3.4 vs. 6.8 days, P < 0.0001) 19 Probable or proven infection (early-onset) PCS Denmark, multicenter Oral switch vs. entirely IV therapy Neonates with clinical signs of infection at 0–72 hours after birth 531 No patient switched to oral therapy was readmitted. (0/478) Median duration of hospitalization was shorter in patients switched to oral therapy. 3.0 days (IQR: 2.5–3.5) vs. 7.4 days (IQR 7.0–7.5) All results are presented as reported in the study.CAP indicates community-acquired pneumonia; CI, confidence interval; CL, confidence level; ED, emergency department; IQR, interquartile range; IV, intravenous; OR, odds ration; PCS, prospective cohort study; RCT, randomized controlled trial; RD, risk difference; RR, relative risk; UTI, urinary tract infection. Oral Antibiotics in Neonatal Infections Neonates with bacterial infections frequently present with nonspecific symptoms and risk of rapid deterioration. While only a minority of neonates with apparent sepsis have culture-confirmed infection, this group of patients is often recommended IV antibiotics for the entire duration of therapy. Uncertainty regarding the efficacy of drug absorption in the first weeks of life and lack of evidence for oral therapy are common concerns about oral switch in neonates. A 2019 systematic review and meta-analysis demonstrated that adequate serum levels of antibiotics can often be achieved after oral administration in neonates.17 Analysis of 31 studies, involving both preterm and term infants, found no evidence to suggest oral antibiotics were associated with worse outcomes in infants treated for bacterial infection. Studies included patients with various suspected and proven bacterial infections, including clinically severe sepsis. Oral antibiotics generally reach maximum concentration later, with lower bioavailability than parenteral dosing, but in the majority achieve adequate serum levels for bacterial killing. In 2022, the same group published the results of a multicenter, open-label, noninferiority trial of early oral antibiotic switch in neonates in the Netherlands with postmenstrual age ≥35 weeks, postnatal age 0–28 days, bodyweight ≥2 kg and probable but unproven bacterial infection. Investigators randomized 504 patients to receive either 7 days of IV antibiotics or initial IV therapy with the oral switch to amoxicillin-clavulanic acid at day 3.18 No significant difference was observed for the primary outcome, bacterial reinfection within 28 days of treatment, nor were there statistically significant differences in the frequency of adverse effects. A subsequent multicenter prospective population-based study of the efficacy of oral switch in neonates with probable bacterial infection published in 2023 reported no significant difference in the rate of readmission at 30 days between infants who were switched to oral antibiotics and those who received a longer duration of IV therapy.19 In 2023, a multicenter cohort study conducted in Denmark investigated whether the oral switch in term neonates with probable or proven early-onset infection was associated with readmission.19 In 531 neonates included for analysis, 90% underwent IV-to-oral antibiotic switch. No patient was readmitted. Furthermore, the median duration of hospitalization was considerably shorter in the patients who were switched to oral antibiotics relative to those who remained on IV therapy (3.0 vs. 7.4 days, respectively). It is increasingly evident that oral switch for clinically stable term neonates with early-onset infection should be considered in routine care. Further research is required, however, for preterm neonates. Although there is growing clinical evidence of the efficacy of early oral switch in probable infection, as well as pharmacological evidence supporting the use of oral antibiotics in neonates and young infants, further research should be performed regarding the clinical efficacy of early oral antibiotic switch in those with various types of proven bacterial infection. Resources currently devoted to extended hospitalization of term neonates to receive IV antibiotics could be reallocated to expanding programs to facilitate safer, early discharge on oral antibiotics. Adult and Pediatric Medicine—What Can We Learn From Each Other? Evidence for the efficacy of oral therapy for infections previously deemed to require IV-only therapy is increasingly strong in adults. There are now multiple published RCTs comparing oral transitional therapy to IV-only therapy for osteomyelitis, both Gram-negative and Gram-positive bacteremia and infective endocarditis.20–22 It is clear that reluctance to adopt the practice of IV-to-oral antibiotic switch has much more to do with tradition and inertia than an absence of data.23 Although literature regarding entirely oral and shorter duration antibiotic therapy in children is not yet as robust as in the adult population, the fact that oral antimicrobial therapy has been established to be safe and effective in multiple concordant RCTs in adult infections, with no evidence to date has demonstrated the superiority of IV-only therapy, demonstrates that IV antibiotic therapy is not somehow magically "more powerful" than oral therapy. Indeed, microbes have no way of knowing by which route antibiotics have been administered. Rather, the salient issue is whether sufficient antibiotics reach the foci of infection such that they are unable to replicate or function. Framed in this manner, and bolstered by trials proving the efficacy of oral therapy, the question is, under what conditions will it be true that antibiotics administered orally will not reach the site of infection at adequate concentrations? Many available data from children parallels the more extensive adult literature that demonstrates sufficient absorption and efficacy of oral antibiotics. This provides reassurance—bidirectionally—that oral therapy is safe and effective for infections, if patients are likely to be capable of taking and absorbing antibiotic therapy administered orally, and agents are available with published data supporting efficacy against the target pathogens. Evidence to support early IV-to-oral antibiotic switch and all-oral therapy of conditions that have conventionally required long courses of IV therapy is needed. Studies investigating bone and joint infections, bacteremia and infections in the immunocompromised host should be a priority for future research. Research directed toward the implementation of existing evidence for shorter therapy and earlier IV-to-oral switch is also needed, including consideration impacts of earlier hospital discharge, reduced healthcare costs and quality-of-life for children. In conclusion, high-quality evidence in both pediatric and adult infections supports the contention that many infections can be treated with oral antibiotics and shorter courses. Clinicians are often reluctant to apply this in practice, despite this evidence. A cultural and imaginative shift is required among those who treat infections—for the benefit of neonates and children.