Introduction

Hemolysis, the rupture of red blood cells leading to the release of their intracellular contents, is a well-recognized phenomenon in laboratory medicine [1, 2]. Within the emergency department (ED), where rapid and accurate laboratory results are critical for timely patient management and interventions, hemolysis poses significant challenges. Hemolyzed samples can introduce inaccuracies in laboratory results, increase turnaround time, and potentially lead to delayed or inappropriate clinical decisions [3–5]. Studies indicate that the prevalence of hemolyzed samples in the ED is >15% for most adults, varying based on the population and laboratory tests under consideration [6–9]. Several factors can instigate hemolysis, including improper phlebotomy techniques, extended tourniquet application, vigorous mixing, and unsatisfactory specimen handling [10, 11]. These factors are especially pertinent to ED practice, where blood is often sampled directly from the peripheral intravenous catheter (PIVC), an established risk factor for hemolysis [12–15]. In busy EDs, it is a common practice to draw blood directly from freshly inserted PIVCs for operational reasons as well as to reduce the number of overall needlesticks. As this approach is unlikely to change, other strategies to reduce ED hemolysis need exploration.

While several studies have investigated hemolysis, most of the evidence is from adult ED patients and the current prevention recommendations are based upon moderate strength evidence at best [2]. Additionally, pediatric-specific data on hemolysis remain scarce, leaving a significant gap in the understanding of the prevalence, risk factors, and clinical implications of hemolysis in pediatric laboratory samples [16, 17]. In pediatric ED visits, where obtaining bloodwork is often routine, venous access is a fundamental procedure [18–20]. Blood sampling and needle procedures are often among the most distressing experiences for pediatric patients, a perception partly attributable to the potential for repeated blood draws. These repeated interventions are recognized as particularly traumatic for both the patients and their caregivers [21–24]. Thus, mitigating hemolysis through targeted interventions could significantly improve care for this vulnerable population.

Identifying the specific factors contributing to hemolysis in pediatric laboratory samples is crucial for the development of interventions to enhance patient care. This manuscript seeks to bridge this knowledge gap, examining the prevalence of hemolyzed samples in the pediatric ED setting, identifying associated risk factors, and discussing potential strategies to reduce hemolysis occurrences. Furthermore, we will assess the impact of hemolyzed samples on patient outcomes within this setting.

Methods

Results

Between January 8^th, 2021, and May 9^th, 2022, a total of 10,462 encounters met inclusion criteria. 1,461 (14.0%) subjects had laboratory evidence of hemolysis, while 9,001 (86.0%) did not. The majority of the cohort was comprised of ages 12–18 (5,222; 49.9%), followed by ages 6–11 (1,878; 18.0%), age 0–1 (1,710; 16.3%), and ages 2–5 (1,652; 15.8%). Female sex comprised 55.3% of the cohort, which was predominantly White (63.6%). The average length of stay was 62.71 hours and 29.4% of all encounters resulted in hospitalization. The most common PIVC sizes were 20 gauge (45.5%) and 22 gauge (42.9%). 34.2% of all PIVCs failed, which was higher amongst the hemolysis cohort (41.3%) compared to the non-hemolysis cohort (32.9%; p<0.001) (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographics and PIVC characteristics of all pediatric patients based on hemolysis.

https://doi.org/10.1371/journal.pone.0299692.t001

In the infant (age 0–1) population, the frequency of hemolysis was 20.1%, with the 24-gauge PIVC being the most common (50.4%) (S1 Table). The frequency of hemolysis was lower among the toddler (2–5) population (17.5%), with the most common PIVC being 22-gauge (93.8%) (S2 Table). Similarly, the 22-gauge (76.9%) was the most common PIVC in children (ages 6–11), with 11.2% having hemolysis (S3 Table). Amongst teenagers, the 20-gauge PIVC was the most common (82.3%), with 11.9% of subjects having hemolysis (S4 Table).

The adjusted multivariable regression analysis, in the toddler subgroup, PIVCs placed in the hand/wrist were associated with higher odds of hemolysis (adjusted odds ratio [aOR] 2.51; 95% confidence interval [CI] 1.88–3.34; p<0.001), as well as the lower extremity (aOR 3.77; 95% CI 1.48–8.99; p = 0.0035). Children had higher odds of hemolysis for PIVCs placed in the hand/wrist (aOR 3.08; 95% CI 2.04–4.58; p<0.001) and forearm (aOR 2.00; 95% CI 1.27–3.06; p = 0.0020) compared to the antecubital fossa. In teenagers, male sex compared to female sex (aOR 1.25; 95% CI 1.04–1.49; p = 0.0149) and Black race compared to White race (aOR 1.40; 95% CI 1.16–1.70; p<0.001) had higher odds of hemolysis. Additionally, compared to PIVCs placed in the antecubital fossa, teenagers had higher odds of hemolysis with PIVCs in the forearm (aOR 1.65; 95% CI 1.31–2.08; p<0.001) and hand/wrist (aOR 2.54; 95% CI 1.92–3.34; p<0.001) (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Multivariate analysis for the odds of hemolysis by age group.

https://doi.org/10.1371/journal.pone.0299692.t002

The odds of failure were assessed via an adjusted multivariate analysis. This did not reveal the presence of hemolysis to significantly impact the odds of failure amongst any of the four age groups. However, in infants, PIVCs placed in the hand/wrist had higher odds of failure (aOR 1.60; 95% CI 1.11–2.33; p = 0.0126) compared to the antecubital fossa. In children, PIVCs placed in the upper arm had lower odds of failure (aOR 0.34; 95% CI 0.11–0.90; p = 0.0419) compared to the antecubital fossa (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Multivariate analysis for the odds of failure by age group.

https://doi.org/10.1371/journal.pone.0299692.t003

Discussion

This study of over 10,000 pediatric patients across a nearly 18-month period revealed an overall hemolysis rate of 14.0%, with the highest proportion occurring in the infant population at 20.1%. This study offers an up-to-date look at hemolysis rates in a large pediatric population. Thus far in pediatrics, the evidence on this topic has been scant, with only two small older studies describing the problem. Bush et al. published a prospective study in 2010 of 221 pediatric patients and found a hemolysis rate of 13%, and in the same year, Berger-Achituv et al. published a prospective study on 40 pediatric patients and identified only one case of hemolysis [16, 17]. Therefore, given the extremely limited prior evidence, in essence this study establishes the baseline for hemolysis in pediatric laboratory samples. Further, unlike the adult population in which similar caliber venous anatomy allows for pooling and analysis as a solitary group, stark venous differences and anatomical considerations based on age must be considered in the pediatric population. Thus, alongside a large sample size, an important strength of this study over prior works is the segmenting of the population into four cohorts to allow for a more robust analysis accounting for the unique anatomical variations among the age groups. Consistent with previous pediatric research, the literature supports such stratification as conventional and appropriate [25–28].

As hemolysis is a frequent occurrence, this could result in care delays and possibly necessitate additional blood draws [29]. As blood sampling in pediatrics is already a difficult and complex procedure, potentially adding needlesticks for additional draws only heightens patient anxiety and discomfort with the ED encounter [30]. Therefore, it is important for clinicians to take steps to reduce this complication. This study identifies risk factors for hemolysis and actionable strategies for prevention. One important consideration is site selection. PIVC’s inserted in the hand or wrist were found to significantly increase the likelihood of hemolysis in pediatric laboratory samples among patients who are at least a year old, mirroring findings from a recent study by the authors in adult populations [9]. Moreover, the correlation between hand/wrist PIVC placements and hemolysis underscores the Infusion Nursing Society’s guidelines to avoid such placements due to the heightened risk of premature failure [31]. Despite the infant population not demonstrating increased odds of hemolysis for hand/wrist PIVCs, those PIVC’s did demonstrate increased odds of failure. Therefore, our results suggest that hand/wrist PIVC placement should not be the primary option for any pediatric age group, especially those facing hospital admission. However, as options for viable insertion sites are often limited in pediatrics, particularly among infants and toddlers, escalation to advanced techniques like illumination or ultrasound guidance may enhance venous visibility in these circumstances [32].

Another important risk factor for hemolysis discovered in this study was that catheter gauge (i.e., catheter diameter) significantly affected the odds of hemolysis in the adolescent population. Higher hemolysis rates were seen with 18-gauge and 22-gauge catheters compared to 20-gauge. This finding is congruent with prior studies on adults, but as most of these investigations were smaller-scale studies and published over two decades ago, the Emergency Nurses Association has not made any firm recommendations regarding catheter diameter and hemolysis in their Clinical Practice Guidelines [2]. Nevertheless, this newer and larger dataset should help substantiate this finding and help shape future clinical guidelines. Interestingly, in contrast to adult studies, hemolysis was not linked with increased PIVC failure rates [9]. A plausible reason might be that pediatric PIVCs typically have shorter dwell times than those in adults. Furthermore, PIVC placement decisions in pediatrics are often made with extra deliberation due to potential adverse effects on both the patient and the caregiver [21–23]. Given the important clinical implications of hemolysis triggering PIVC failure, further research on this topic is warranted in the pediatric population.

Limitations

While the investigation yields insightful findings, it is important to acknowledge certain limitations that potentially impact the results and interpretations. First, the research is subject to potential selection bias due to the non-random selection of participants. The use of convenience sampling could limit the generalizability of the findings to the broader pediatric population. However, a very large sample size likely mitigates this limitation. Second, the retrospective study design inherently limits the capacity to control for unobserved confounding variables. Consequently, the observed associations should not be construed as cause-and-effect relationships. Third, the study adopted a stratification of the pediatric population into four cohorts based on age-related venous and anatomical differences. While beneficial for a nuanced and clinically relevant analysis, this approach restricts the sample size in each cohort, which may not be powered to detect effects when they may exist. Lastly, this study did not assess the precise impact on delays in care, rate of redraw, or medical errors due to hemolysis. This aspect underscores the need for further research to fully understand the implications of our findings in the context of real-world pediatric clinical practice.

Conclusions

Hemolysis in the pediatric population appears to occur frequently and at a similar rate to the adult population. Compared to PIVCs placed in the antecubital fossa, PIVCs in the hand/wrist demonstrate higher odds of hemolysis, and clinicians should consider placing PIVCs in alternative locations if clinically appropriate. Further studies are needed to better understand the clinical implications of pediatric hemolysis.

Supporting information