SGEM #470: Here We Go Up Up Up or Lateral for Infant Lumbar Punctures

Reference: Pessano S, et al. Positioning for lumbar puncture in newborn infants. Cochrane Database Syst Rev. December 2023 Date: February 7, 2025 Dr. Lauren Rosenfeld Guest Skeptic: Dr. Lauren Rosenfeld is a PGY-3 emergency medicine resident at George Washington University. She is also a new podcast host for Emergency Medicine Residents’ Association (EMRA) Cast Series. Case: A five-day-old girl is brought to the emergency department (ED) for fever by her parents. She was born full-term and seemed to be doing very well after the family returned home. Her mother had an uneventful pregnancy and delivery. Today, the parents thought she was feeling warm and took the girl’s temperature, which was 101°F (38.3°C). They called the pediatrician, who told them to go to the ED for more testing and warned them of the likelihood that their baby may need a lumbar puncture. The worried father asks you, “What is a lumbar puncture? Will it hurt?” Her mother asks you, “Is it like when I got an epidural before delivering? Will you sit her up for it? She can’t sit yet.”  Background:  We have covered the topic of febrile infants and lumbar punctures (LP) before on the SGEM. However, we typically focused on the febrile infant part. Today we’re going to talk more about performing the procedure of a lumbar puncture on babies.  In the ED, lumbar punctures are typically performed in infants with fever in the evaluation for invasive bacterial infections including meningitis. There are many thoughts and bits of advice around how to perform an LP including the proper position, when to remove the stylet from the needle, what kind of analgesia to use, etc. There are multiple positions to set up the lumbar puncture. Commonly, patients can be placed on their side in the lateral decubitus, bend the neck so the chin is close to the chest, hunch the back, and bring the knees toward the chest to approximate the fetal position. Alternatively, patients may also sit upright and then bend their head and shoulders forward. When it comes to infants, most of the time, we are relying on someone else to help hold the baby in those positions as we’re performing the LP. Sometimes, these babies can have episodes of oxygen desaturation when they get held in that position for too long. Clinical Question: How does the positioning of infants during lumbar puncture (lateral decubitus vs sitting vs prone) affect success rates and adverse events?  Reference: Pessano S, et al. Positioning for lumbar puncture in newborn infants. Cochrane Database Syst Rev. December 2023 Population: preterm and term infants of postmenstrual age up to 46 weeks and 0 days. Age 4.9 hours to 5 weeks Intervention: Infants positioned in a lateral decubitus position. Comparison: Infants positioned in a sitting position or prone position​. Outcome: Primary Outcome(s): Successful lumbar puncture on the first attempt, with < 500 red blood cells/mm3. Total number of lumbar puncture attempts (successful or unsuccessful). Episodes of bradycardia, defined as a decrease in HR of more than 30% below baseline or less than 100bpm for 10 seconds or longer. Secondary Outcomes: Time to perform LP, episodes of desaturation (SpO2 <80%), apnea, need for pain/sedation medication, skin changes at LP site, infection rate related to LP, pain, and parental satisfaction. Type of Study: Systematic Review Meta-analysis Authors’ Conclusions: “When compared to sitting position, lateral decubitus position probably results in little to no difference in successful lumbar puncture procedure at first attempt. None of the included studies reported the total number of lumbar puncture attempts as specified in this review. Furthermore, infants in a sitting position likely experience less episodes of bradycardia and oxygen desaturation than in the lateral decubitus, and there may be little to no difference in episodes of apnea. Lateral decubitus position results in little to no difference in time to perform the lumbar puncture compared to sitting position. Pain intensity during and after the procedure was reported using a pain scale that was not included in our prespecified tools for pain assessment due to its high risk of bias. Most study participants were term newborns, thereby limiting the applicability of these results to preterm babies. When compared to prone position, lateral decubitus position may reduce successful lumbar puncture procedure at first attempt. Only one study reported on this comparison and did not evaluate adverse effects. Further research exploring harms and benefits and the effect on patients' pain experience of different positions during lumbar puncture using validated pain scoring tools may increase the level of confidence in our conclusions.” Quality Checklist for Therapeutic Systematic Reviews: The clinical question is sensible and answerable. Yes The search for studies was detailed and exhaustive. Yes The primary studies were of high methodological quality. No The assessment of studies were reproducible. Yes The outcomes were clinically relevant. Yes There was low statistical heterogeneity for the primary outcomes. No  The treatment effect was large enough and precise enough to be clinically significant. No Financial Conflicts of Interest: None Results: They included five studies with 1,476 participants. The mean gestational age of the infants in the included studies ranged from 31 weeks to 41 weeks, with the largest study enrolling mostly term newborns. The mean postnatal age at the time of procedure completion ranged from 4.9 hours to five weeks​. Key Results: There was not much difference in LP success with the lateral decubitus position compared to other positions. However, lateral decubitus positioning may be associated with more episodes of bradycardia and desaturations. Primary Outcomes: No difference in LP success between lateral decubitus and sitting position (RR 0.99, 95% CI 0.88-1.12). Lateral decubitus positioning did increase episodes of bradycardia (RR 1.72, 95% CI 1.08 to 2.76). That was a number needed to harm of 33. Lateral decubitus positioning also increased episodes of desaturation (RR 2.1, 95% CI 1.42 to 3.08). That was a number needed to harm of 17. Secondary Outcomes: Key patient-centered outcomes such as pain, infection risk, sedation needs, and parental satisfaction remain unreported. Included Studies Overall, they only found five studies to include in this review. Four were randomized controlled trials, and one was a quasi-randomized controlled trial. Most of the data for this review came from one study that had 1082 participants which was around 73% of all the participants included in the review. When we look at the outcomes they were trying to assess, most of the time only 2 or 3 studies reported the outcome of interest, making testing for heterogeneity challenging Certainty of Evidence Even though they included five studies, these studies only included a total of 1,476 patients. Because of the limited data, many of the outcomes they were looking at were moderate or low certainty of evidence. There was only one outcome that achieved high certainty which was that there was little to no difference in time to perform lumbar puncture when comparing lateral decubitus to sitting position. The time to perform a lumbar puncture may vary quite a bit depending on experience of the clinician performing the procedure. It may be the case that the longer it takes to perform the LP, the more risk of adverse events like desaturations or bradycardia occurs because the baby is scrunched up in that position. Patient-Oriented vs. Monitor Oriented Outcomes Their outcomes of interest for a mix of patient-oriented outcomes (POOs) and monitor-oriented outcomes (MOOs). One fairly important patient-oriented outcome that wasn’t reported across the studies included was the number of LP attempts. I would say that as a parent and caregiver, this is important. I don’t know how happy I would be if someone was “successful” with their LP but in the process, they turned the baby into a pin cushion. The outcomes of desaturations and bradycardia are monitor-oriented outcomes. The definition for what counted as a desaturation or bradycardic episode varied or was not reported across the studies included. The authors defined desaturation as pulse oximetry <80% with no minimum duration and apnea as interruption in breathing for more than 20 seconds. It is unclear if these desaturations or bradycardic episodes were sustained, self-resolving, required intervention. Is it accurate to attribute these events to the LP procedure itself? Indications for Lumbar Puncture There was variation in the populations that were being studied in each of the included studies. One study included sick neonates. One study included infants 1 to 90 days undergoing LP in the emergency department. It did not specify the indications. One study included preterm infants who received LP for spinal anesthesia before inguinal hernia repair. The largest study included infants 27 to 44 weeks corrected gestational age. Most of these study participants were included due to concerns for infection or sepsis. The difference in the included populations could have also impacted the results. For example, it’s possible that sicker babies may be more at risk of having episodes of desaturations, bradycardia, or apnea compared to those who were either well-appearing febrile infants or receiving LP for anesthesia. Unmeasured/Unreported Confounders There are a lot of factors to consider when performing a lumbar puncture. That can include adequate analgesia, technique of the person holding the baby, early stylet removal, and experience of the performing physician [1-4].