When to start and stop caffeine and why respiratory status matters

https://doi.org/10.1016/j.siny.2020.101175Get rights and content

Abstract

Caffeine as tested in the CAP trial is safe and effective for preterm infants with birthweights less than 1250 g. Evidence for its use beyond the indications and timing used in this trial is of low quality and conflicting. Some studies suggest that earlier use of caffeine increases the risk of mortality while others suggest it has important benefits. It appears that infants with apnea of prematurity and those receiving assisted ventilation are the most likely to benefit from caffeine. Based on currently available evidence, routine early prescription of caffeine does not appear to be indicated. Infants continue to have potentially damaging episodes of hypoxia secondary to apnea beyond 34 weeks’ postmenstrual age. It is unclear whether prolonged use of caffeine improves outcomes in these infants. Randomized trials to resolve these uncertainties are required. They need to be large, at least the size of the CAP trial, and include neurodevelopmental outcomes.

Section snippets

Evidence of caffeine's potential to cause harm as well as benefit

It is important to note that the CAP trial was initially conceived as a trial of safety of a drug class that was already in widespread use [5]. Although there was evidence of efficacy in terms of an ability to reduce the frequency of apnea, only 222 preterm infants had been treated with methylxanthines in the context a of randomized trial [5]. Both the number of patients treated and the duration of follow-up provided no reassurance of either short or long-term safety.

At the time the CAP trial

Further evidence from the CAP trial

Together, indirect animal and human data raise questions about whether the results of the CAP trial provide grounds for more aggressive use, particularly whether caffeine should be used prophylactically in the early hours or days of life and whether it should be continued beyond the post-menstrual ages (PMAs) reported in the trial. CAP was a pragmatic trial, designed to represent usual clinical practice. Clinicians prescribed the study drug when they thought infants were “candidates for

Evidence from cohort studies

At least 14 cohort studies have addressed the question of early versus late initiation of caffeine therapy, all of which might be considered as being at serious or critical risk of bias [19]. Two publications from the Canadian Neonatal Network addressed the short and long term differences in outcomes following early (within the first two days after birth) and late (on or after the third day). In a cohort recruited in 2010–2012 they reported that infants receiving caffeine early had shorter

Systematic reviews of cohort studies

At least four systematic reviews have been conducted addressing this issue [19,[29], [30], [31]]. Pooling data from five cohort studies, three suggested that early (before three days) caffeine use was associated with a reduced risk of BPD [[29], [30], [31]]. Interestingly, one noted an increased mortality rate in the early treated infants [30], one did not perform a pooled analysis but noted the increased mortality rate in the largest trial [31], and the other suggested that early caffeine

Randomized trials

Two randomized trials have addressed the issue of very early commencement of caffeine [33]. In a trial of 30 infants, Dekker et al. compared the strategy of administering caffeine to newly born infants 24–30 weeks' gestation in the delivery room via the umbilical vein to therapy starting after arrival in the NICU [33]. They found that caffeine increased respiratory effort in the first minutes of life but the study was not powered to evaluate important in-hospital outcomes. Although interesting,

When to stop caffeine therapy

The time to reach various physiological milestones was assessed in a cohort of infants born ≤32 weeks' gestation in the Northern California Kaiser Permanente Medicare Program [36]. The median (interquartile range) PMA of infants' last day of methylxanthine therapy was 33.7 (32.9–35.1) weeks', last bradycardia spell 34.3 (33.0–35.9) weeks' and last apnea spell 33.9 (32.7–35.4) weeks'. On average, infants born before 27 weeks' gestation achieved these milestones at least a week later than their

Why respiratory status matters

The ‘treatment creep’ seen in the prescription of caffeine would not be a problem if the drug had no toxicity. However, as noted caffeine has broad multisystem effects and there is evidence of potential harm from both human and animal studies.

Importantly, the CAP trial showed that about one-half of the neuroprotective effect of caffeine seen at 18 months was explained by the earlier discontinuation of any form of positive pressure ventilation [3]. In another subgroup analysis, we showed that

Conclusions and recommendations

Expert opinion is usually represented as the lowest tier in the evidence pyramid. However, some point out that it forms the foundation on which higher forms of evidence rest. The CAP trial demonstrated that overall, caffeine is safe and beneficial for extremely preterm infants. Bancalari reminds us of the many “therapeutic misadventures” seen in neonatology and the fact that caffeine is a drug with broad multi-system effects [43]. He suggests that caffeine is not the silver bullet we are

Acknowledgments

I would like to acknowledge support through an Australian National Health and Medical Research Council Practioner Fellowship grant (#1157782).

References (44)

  • E. Bancalari

    Current management of apnea in premature infants: is caffeine the magic bullet?

    Early Hum Dev

    (2014)
  • B. Schmidt et al.

    Caffeine therapy for apnea of prematurity

    N Engl J Med

    (2006)
  • B. Schmidt et al.

    Long-term effects of caffeine therapy for apnea of prematurity

    N Engl J Med

    (2007)
  • J.H. Thurston et al.

    Aminophylline increases cerebral metabolic rate and decreases anoxic survival in young mice

    Science

    (1978)
  • M. Yazdani et al.

    Chronic caffeine intake alters the composition of various parts of the brain in young growing rats

    Dev Pharmacol Ther

    (1988)
  • S. Endesfelder et al.

    Prevention of oxygen-induced inflammatory lung injury by caffeine in neonatal rats

    Oxid Med Cell Longev

    (2020)
  • S. Endesfelder et al.

    Antioxidative effects of caffeine in a hyperoxia-based rat model of bronchopulmonary dysplasia

    Respir Res

    (2019)
  • E. Ali et al.

    Caffeine is a risk factor for osteopenia of prematurity in preterm infants: a cohort study

    BMC Pediatr

    (2018)
  • J. Hwang et al.

    Hemodynamic effects on systemic blood flow and ductal shunting flow after loading dose of intravenous caffeine in preterm infants according to the patency of ductus arteriosus

    J Kor Med Sci

    (2018)
  • M.A. Abdel Wahed et al.

    Effect of caffeine on superior mesenteric artery blood flow velocities in preterm neonates

    J Matern Fetal Neonatal Med

    (2019)
  • J. Huvanandana et al.

    Cardiovascular impact of intravenous caffeine in preterm infants

    Acta Paediatr

    (2019)
  • S.W. Lagakos

    The challenge of subgroup analyses--reporting without distorting

    N Engl J Med

    (2006)

    • Cited by (10)

    View all citing articles on Scopus
    View full text
    Crown Copyright © 2020 Published by Elsevier Ltd. All rights reserved.