TITLE OF WRITE-UP
Early Switching from Intravenous to Oral Antibiotics in Neonates with Probable Bacterial Infection
MANUSCRIPT CITATION
Keij FM, Kornelisse RF, Hartwig NG, van der Sluijs-Bens J, van Beek RHT, van Driel A, van Rooij LGM, van Dalen-Vink I, Driessen GJA, Kenter S, von Lindern JS, Eijkemans M, Stam-Stigter GM, Qi H, van den Berg MM, Baartmans MGA, van der Meer-Kappelle LH, Meijssen CB, Norbruis OF, Heidema J, van Rossem MC, den Butter PCP, Allegaert K, Reiss IKM, Tramper-Stranders GA. Efficacy and safety of switching from intravenous to oral antibiotics (amoxicillin-clavulanic acid) versus a full course of intravenous antibiotics in neonates with probable bacterial infection (RAIN): a multicentre, randomised, open-label, non-inferiority trial. Lancet Child Adolesc Health. 2022 Nov;6(11):799-809. doi: 10.1016/S2352-4642(22)00245-0. Epub 2022 Sep 9. PMID: 36088952.
REVIEWED BY
Nehad Nasef
Professor of Pediatrics,
Department of Pediatrics/Neonatology, Mansoura University Children’s Hospital, Mansoura, Egypt.
Islam Nour
Associate Professor of Pediatrics,
Department of Pediatrics/Neonatology, Mansoura University Children’s Hospital, Mansoura, Egypt.
Hesham Abdel-Hady
Professor of Pediatrics,
Department of Pediatrics/Neonatology, Mansoura University Children’s Hospital, Mansoura, Egypt.
FUNDING
None
CONFLICTS OF INTEREST
None
TYPE OF INVESTIGATION
Treatment
QUESTION
(P) In clinically well neonates born at or after 35 completed weeks’ gestation with probable bacterial infection, (I) is intravenous-to-oral antibiotic switch policy, at least as effective as, (C) a full course of intravenous antibiotics, (O) in reducing bacterial (re) infection within 28 days after finishing antimicrobial therapy (T)?
METHODS
- Design: Randomized controlled, open label, multicenter trial, 17 large teaching hospitals in the Netherlands.
- Allocation: Block randomization of eligible neonates by (1:1) ratio was applied using a web-based computer-generated program, with block sizes of 2 and 4. The randomization was stratified by study center.
- Blinding: It was not feasible to mask caregivers and parents; the study statistician was masked to treatment allocation.
- Follow-up period: Up to 28 days after treatment completion.
- Setting: 17 large teaching hospitals, represented in 19 recruitment sites, in the Netherlands between February 2018 and May 2021.
- Patients:
- Inclusion criteria: Clinically well neonates delivered at or after 35 completed weeks’ gestation with birth weight ≥ 2.00 kg, who fulfilled criteria of probable bacterial infection for which empiric antibiotic therapy was prescribed for 7 days. Eligibility criteria were: (1) Evidence of probable bacterial infection as following: maternal risk factors; and/or clinical symptoms; and elevated inflammatory markers (C reactive protein or procalcitonin) (2) Negative blood culture.
- Exclusion criteria: All neonates with culture-proven bloodstream bacterial infection, severe localized infection (e.g. meningitis, osteomyelitis, nescrotising enterocolitis), severe clinical sepsis necessitating intensive care treatment, continuous need for a central venous catheter, severe hyperbilirubemia beyond exchange transfusion threshold. Besides, infants of parents who were unable to administer medications were considered ineligible.
- Intervention:
- In the intervention group, intravenous antibiotics had been replaced by eight- hourly dosing regimen of 25/6.25 mg/kg of oral amoxicillin/ clavulanic acid, after 48 to 72 hours.
- In the control group, the full course of intravenous antibiotics had been accomplished.
- In both groups, neonates will be treated for a total of 7 days.
- Outcomes:
- Primary outcome: Cumulative bacterial re-infection rate within 28 days after finishing of antibacterial therapy, assessed by the local investigator or research nurse at the outpatient department. Bacterial re-infection was identified by clinical symptoms of infection and fever (>38.0°C) or hypothermia (<36.0°C) and elevated inflammatory parameters with a need for prolonged antimicrobial therapy for more than 48–72 hours.
- Secondary outcomes: clinical deterioration within 7 days after randomization, duration of hospitalization, readmission rate, quality of life and adverse events. The quality of life had been assessed by parents’ questionnaire on day 7and day 21, in addition to outpatient clinic visit on day 35 to check the following items: painful procedures; breastfeeding; sleep quality; gastrointestinal symptoms; and number of intravenous catheter reinsertions.
- Analysis and Sample Size:
- The authors assumed a 1% reinfection rate after intravenous antibiotics completion. The sample size of 231 infants per group was estimated considering a noninferiority margin for intravenous-to-oral antibiotic switch strategy of 3% above the re-infection rate with applying the full-course intravenous antibiotic policy, at a one-sided α level of 0⋅025, and a power of 90%. In addition, with assumption of 10% dropout rate, the required total sample size of 510 infants was computed.
- Both and intention-to-treat and per-protocol analysis had been performed. Regarding primary and secondary outcomes, the statistical significance threshold was set at a p-value less than 0.025 and 0.05, respectively.
- 95% Wilson confidence interval was estimated for intergroup differences.
- Patient follow-up:
Of 829 infants assessed for eligibility, 510 were recruited (255 per group). Eventually, Intention to treat analysis were performed for 252 neonates in each group, whereas, data of 246 and 203 infants were analyzed as per-protocol in the oral amoxicillin–clavulanic acid group and the intravenous antibiotics group, respectively.
MAIN RESULTS:
- The median and interquartile range of gestational age of recruited neonates was 40⋅3 weeks (39·3–41·0).
- Respiratory support was initiated within first 48 hours of life for 196 (39%) of 504 neonates of the whole cohort.
- Neonates in the intravenous antibiotics group had a significantly higher rates of protocol violations [53 (21%) vs. 18 (7%), P = <0·0001] compared with the amoxicillin–clavulanic acid group.
- The leading cause of protocol violations in the intravenous antibiotics group was venous access difficulty as reported in 49 out of 245 patients (20%).
- The primary outcome of the cumulative reinfection rate at day 28 was reported in 1 of 252 neonates (<1%) in the amoxicillin–clavulanic acid group and 1 of 252 neonates (<1%) in the intravenous antibiotics group, between groups difference 0 [95% CI –1⋅9 to 1⋅9]; p non-inferiority <0⋅0001).
- Neonates in the in the amoxicillin–clavulanic acid group had significantly shorter duration of hospital stay [3.4 (3·0 to 4·1) vs. 6·8 (6·5 to 7·0), P = <0·0001] compared with than the intravenous antibiotics group.
- No statistically significant differences between both arms as regards adverse events either per intention-to -treat analysis (127 [50%] vs 113 [45%]; p=0·247) or per-protocol analysis (129 [52%] vs 82 [40%]; p= 0·074).
- The re-admission rates were similar between both groups.
- No statistically significant difference between both arms regarding the readmission rates and the quality of life assessment items.
CONCLUSION:
The authors conclude that intravenous to- oral switch policy is not inferior to a full course of intravenous antibiotics completion in neonates with probable bacterial infection without increased the incidence of adverse events.
COMMENTARY
Diagnosis of early-onset neonatal sepsis depends mainly on the presence of perinatal risk factors, clinical signs, inflammatory markers, and isolation of organisms from blood culture or sterile body fluids. The current standard of care recommends treating neonates with culture-confirmed early-onset sepsis by intravenous antibiotics for 5 -7 days (5) and discontinuing an antibiotic when blood culture is sterile at 36 to 48 hours of incubation unless there is evidence of site-specific infection (6). Although culture-confirmed sepsis represents 0.4 – 0.8 / 1000 of term infants born in high-income countries (3), evidence has shown that neonates receive systemic antibiotics at a higher rate for cases of culture-negative probable bacterial infection (1, 7) which expose them to longer NICU stay, interferes with parent-child bonding, and increases hospital-related risks and substantial costs (4). The RAIN study investigated the value of an early intravenous-to-oral antibiotic switch after 48 to 72 hours compared to the standard full course of intravenous antibiotics in reducing bacterial reinfection in clinically stable term and late preterm neonates diagnosed with a probable bacterial infection. The authors found that an early intravenous-to-oral antibiotic switch was associated with a similar reinfection rate within 28 days after finishing antimicrobial therapy, a shorter length of hospital stay, a non-significant rate of adverse events, and a similar rate of readmission compared to a full course of intravenous antibiotics (2).
The study is strengthened by being the first to address the probability of using oral antibiotics in neonates in a high-income setting, multicenter with a high compliance rate to follow-up, and the use of a pragmatic approach to discharge allowing for evaluation of the safety of home-based care. The authors acknowledged their limitations of recruiting stable infants with a diagnosis of probable bacterial infection which has no clear clinical standards or consistent cutoff values of diagnostic biomarkers, use of nonspecific clinical signs for diagnosis which may exist with the transition from intra-uterine to extra-uterine life, being an open-label study which may have influenced the observations during the treatment course and follow-up, and the use of oral amoxicillin-clavulanic acid which lacks the coverage for resistant E-coli strain with a possible effect on the developing gut microbiome (2).
Although the current study suggested that the administration of oral antibiotics in a home-based setting is safe in infants with a probable bacterial infection in high-income countries, physicians are challenged with the sensitivity and specificity of the tools used for the diagnosis of probable bacterial infection. In the current study, 20% and 13% of the recruited infants to the oral and intravenous antibiotic groups respectively had no clinical symptoms at the onset of the study which requires only enhanced observation without cultures or antibiotics according to the current standards of care (6). Respiratory symptoms accounted for 83% and 79% of the clinical presentation in the oral and intravenous antibiotic groups respectively, 39% of them required respiratory support in the first 48 hours of life, which raises the possibility of transient tachypnea of newborns particularly in presence of rapid improvement of respiratory symptoms within 48 hours. Moreover, the authors used an arbitrary value of CRP or procalcitonin as biomarkers for probable infection which are neither sensitive nor specific to guide early onset sepsis decision. Both CRP and procalcitonin concentrations increase in response to a variety of inflammatory stimuli rather than infection and procalcitonin concentrations also increase naturally over the first 36 hours of life (8). Furthermore, the authors reported adverse events in 53% of the included infants in the oral antibiotics group which makes the value of extending antibiotic therapy beyond the first 48 hours of life questionable in absence of clear criteria for diagnosis.
In conclusion, the RAIN study provides a safe alternative of oral antibiotics therapy at home for infants with a probable bacterial infection which shortens the hospital stay and enhances family-infant bonding. However, evidence is still needed on the proper tools for diagnosis of probable bacterial infection to avoid unnecessary antibiotics overuse.
REFERENCES
-
Fjalstad JW, Stensvold HJ, Bergseng H, Simonsen GS, Salvesen B, Ronnestad AE, Klingenberg C (2016) Early-onset Sepsis and Antibiotic Exposure in Term Infants: A Nationwide Population-based Study in Norway. Pediatr Infect Dis J 35:1-6
-
Keij FM, Kornelisse RF, Hartwig NG, van der Sluijs-Bens J, van Beek RHT, van Driel A, van Rooij LGM, van Dalen-Vink I, Driessen GJA, Kenter S, von Lindern JS, Eijkemans M, Stam-Stigter GM, Qi H, van den Berg MM, Baartmans MGA, van der Meer-Kappelle LH, Meijssen CB, Norbruis OF, Heidema J, van Rossem MC, den Butter PCP, Allegaert K, Reiss IKM, Tramper-Stranders GA (2022) Efficacy and safety of switching from intravenous to oral antibiotics (amoxicillin-clavulanic acid) versus a full course of intravenous antibiotics in neonates with probable bacterial infection (RAIN): a multicentre, randomised, open-label, non-inferiority trial. Lancet Child Adolesc Health 6:799-809
-
Nanduri SA, Petit S, Smelser C, Apostol M, Alden NB, Harrison LH, Lynfield R, Vagnone PS, Burzlaff K, Spina NL, Dufort EM, Schaffner W, Thomas AR, Farley MM, Jain JH, Pondo T, McGee L, Beall BW, Schrag SJ (2019) Epidemiology of Invasive Early-Onset and Late-Onset Group B Streptococcal Disease in the United States, 2006 to 2015: Multistate Laboratory and Population-Based Surveillance. JAMA Pediatr 173:224-233
-
Patel SJ, Saiman L (2010) Antibiotic resistance in neonatal intensive care unit pathogens: mechanisms, clinical impact, and prevention including antibiotic stewardship. Clin Perinatol 37:547-563
-
Paul SP, Khattak H, Kini PK, Heaton PA, Goel N (2022) NICE guideline review: neonatal infection: antibiotics for prevention and treatment (NG195). Arch Dis Child Educ Pract Ed 107:292-297
-
Puopolo KM, Benitz WE, Zaoutis TE (2018) Management of Neonates Born at >/=35 0/7 Weeks’ Gestation With Suspected or Proven Early-Onset Bacterial Sepsis. Pediatrics 142
-
Stocker M, van Herk W, El Helou S, Dutta S, Fontana MS, Schuerman F, van den Tooren-de Groot RK, Wieringa JW, Janota J, van der Meer-Kappelle LH, Moonen R, Sie SD, de Vries E, Donker AE, Zimmerman U, Schlapbach LJ, de Mol AC, Hoffman-Haringsma A, Roy M, Tomaske M, Kornelisse RF, van Gijsel J, Visser EG, Willemsen SP, van Rossum AMC (2017) Procalcitonin-guided decision making for duration of antibiotic therapy in neonates with suspected early-onset sepsis: a multicentre, randomised controlled trial (NeoPIns). Lancet 390:871-881
-
Su H, Chang SS, Han CM, Wu KY, Li MC, Huang CY, Lee CL, Wu JY, Lee CC (2014) Inflammatory markers in cord blood or maternal serum for early detection of neonatal sepsis-a systemic review and meta-analysis. J Perinatol 34:268-274