Painkillers other than opioids to treat pain in babies undergoing painful procedures

Key messages 

• We did not find enough evidence on painkillers that are not opioids to manage pain in babies undergoing painful procedures. We found only two small studies that compared a painkiller (ketamine) with either another painkiller (an opioid) or sweet solution to manage different procedures. 

• Larger studies on a variety of painkillers are needed to provide a better understanding of the benefits and harms of the different painkillers and the best way to give them. 

What did we want to find out? 

Babies, particularly those born too early or sickest, may undergo many painful procedures during their hospital stay. It is still unclear which painkillers are best for adequate and safe pain relief. We specifically focused on studies evaluating non-steroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen, and N-methyl-D-aspartate (NMDA) receptor antagonists, such as ketamine, for babies during painful procedures. We wanted to find out how they affected pain intensity in babies during the procedures and any side effects the painkillers caused. 

What did we do? 

We searched for studies that compared: 

• NMDA receptor antagonists (e.g. ketamine) or NSAIDs (e.g. ibuprofen) versus no treatment, placebo (dummy treatment), oral sweet solution, or non-painkiller intervention;

• one painkiller versus another painkiller; or 

• different ways of giving the same painkillers (e.g. through the mouth or vein). 

We evaluated the studies and rated our confidence in the evidence based on study methods. 

What did we find? 

We found two studies involving a total of 269 babies undergoing painful procedures. One study conducted in Nigeria compared giving ketamine versus sugar syrup through the mouth for circumcision. The other study, conducted in India, compared fentanyl and ketamine given through the vein during laser treatment for an eye disease. In both studies, the babies who received ketamine had lower pain scores; however, the studies used different methods and evaluated different procedures, making it difficult to draw any conclusions. The latter study also provided unclear evidence on the effect of the treatments on breathing and blood pressure problems. 

What are the limitations of the evidence? 

We are not confident in the evidence and are very uncertain about the results due to the limited studies included, their small size, and their use of methods likely to produce errors. 

How up-to-date is this evidence? 

The evidence is current to June 2022. 

Authors' conclusions: 

The two small included studies comparing ketamine versus either placebo or fentanyl, with very low-certainty evidence, rendered us unable to draw meaningful conclusions. The evidence is very uncertain about the effect of ketamine on pain score during the procedure compared with placebo or fentanyl. We found no evidence on NSAIDs or studies comparing different routes of administration.

Future research should prioritize large studies evaluating non-opioid analgesics in this population. As the studies included in this review suggest potential positive effects of ketamine administration, studies evaluating ketamine are of interest. Furthermore, as we identified no studies on NSAIDs, which are widely used in older infants, or comparing different routes of administration, such studies should be a priority going forward. 

Read the full abstract...

Neonates are an extremely vulnerable patient population, with 6% to 9% admitted to the neonatal intensive care unit (NICU) following birth. Neonates admitted to the NICU will undergo multiple painful procedures per day throughout their stay. There is increasing evidence that frequent and repetitive exposure to painful stimuli is associated with poorer outcomes later in life.

To date, a wide variety of pain control mechanisms have been developed and implemented to address procedural pain in neonates. This review focused on non-opioid analgesics, specifically non-steroidal anti-inflammatory drugs (NSAIDs) and N-methyl-D-aspartate (NMDA) receptor antagonists, which alleviate pain through inhibiting cellular pathways to achieve analgesia. 

The analgesics considered in this review show potential for pain relief in clinical practice; however, an evidence summation compiling the individual drugs they comprise and outlining the benefits and harms of their administration is lacking. We therefore sought to summarize the evidence on the level of pain experienced by neonates both during and following procedures; relevant drug-related adverse events, namely episodes of apnea, desaturation, bradycardia, and hypotension; and the effects of combinations of drugs. 

As the field of neonatal procedural pain management is constantly evolving, this review aimed to ascertain the scope of non-opioid analgesics for neonatal procedural pain to provide an overview of the options available to better inform evidence-based clinical practice. 


To determine the effects of non-opioid analgesics in neonates (term or preterm) exposed to procedural pain compared to placebo or no drug, non-pharmacological intervention, other analgesics, or different routes of administration.

Search strategy: 

We searched the Cochrane Library (CENTRAL), PubMed, Embase, and two trial registries in June 2022. We screened the reference lists of included studies for studies not identified by the database searches.

Selection criteria: 

We included all randomized controlled trials (RCTs), quasi-RCTs, and cluster-RCTs in neonates (term or preterm) undergoing painful procedures comparing NSAIDs and NMDA receptor antagonists to placebo or no drug, non-pharmacological intervention, other analgesics, or different routes of administration. 

Data collection and analysis: 

We used standard Cochrane methods. Our main outcomes were pain assessed during the procedure and up to 10 minutes after the procedure with a validated scale; episodes of bradycardia; episodes of apnea; and hypotension requiring medical therapy.

Main results: 

We included two RCTs involving a total of 269 neonates conducted in Nigeria and India. 

NMDA receptor antagonists versus no treatment, placebo, oral sweet solution, or non-pharmacological intervention

One RCT evaluated using oral ketamine (10 mg/kg body weight) versus sugar syrup (66.7% w/w at 1 mL/kg body weight) for neonatal circumcision. 

The evidence is very uncertain about the effect of ketamine on pain score during the procedure, assessed with the Neonatal Infant Pain Scale (NIPS), compared with placebo (mean difference (MD) −0.95, 95% confidence interval (CI) −1.32 to −0.58; 1 RCT; 145 participants; very low-certainty evidence). No other outcomes of interest were reported on.

Head-to-head comparison of different analgesics

One RCT evaluated using intravenous fentanyl versus intravenous ketamine during laser photocoagulation for retinopathy of prematurity. Neonates receiving ketamine followed an initial regimen (0.5 mg/kg bolus 1 minute before procedure) or a revised regimen (additional intermittent bolus doses of 0.5 mg/kg every 10 minutes up to a maximum of 2 mg/kg), while those receiving fentanyl followed either an initial regimen (2 μg/kg over 5 minutes, 15 minutes before the procedure, followed by 1 μg/kg/hour as a continuous infusion) or a revised regimen (titration of 0.5 μg/kg/hour every 15 minutes to a maximum of 3 μg/kg/hour). The evidence is very uncertain about the effect of ketamine compared with fentanyl on pain score assessed with the Premature Infant Pain Profile-Revised (PIPP-R) scores during the procedure (MD 0.98, 95% CI 0.75 to 1.20; 1 RCT; 124 participants; very low-certainty evidence); on episodes of apnea occurring during the procedure (risk ratio (RR) 0.31, 95% CI 0.08 to 1.18; risk difference (RD) −0.09, 95% CI −0.19 to 0.00; 1 study; 124 infants; very low-certainty evidence); and on hypotension requiring medical therapy occurring during the procedure (RR 5.53, 95% CI 0.27 to 112.30; RD 0.03, 95% CI −0.03 to 0.10; 1 study; 124 infants; very low-certainty evidence). The included study did not report pain score assessed up to 10 minutes after the procedure or episodes of bradycardia occurring during the procedure.

We did not identify any studies comparing NSAIDs versus no treatment, placebo, oral sweet solution, or non-pharmacological intervention or different routes of administration of the same analgesics. We identified three studies awaiting classification.