Introduction: Establishing peripheral vein access is challenging for pediatric residents and a painful procedure for neonates. We assessed the efficacy of a red light-emitting diode transilluminator during peripheral vein catheter insertion performed by pediatric residents. Methods: Patients were stratified by current weight (≤1,500 g, >1,500 g) and randomized to the transillumination or the control group. The first three attempts were performed by pediatric residents, followed by three attempts by a neonatologist. The primary outcome was success at first attempt. Secondary comparisons included time to successful insertion and overall success rates of residents and neonatologists. Results: A total of 559 procedures were analyzed. The success rate at resident’s first attempt was 44/93 (47%) with transillumination versus 44/90 (49%) without transillumination (p = 0.88) in the strata ≤1,500 g and 103/188 (55%) with transillumination versus 64/188 (34%) without transillumination in the strata >1,500 g (p < 0.001). The overall success rate for residents was 86% in the transillumination versus 73% in the control group in the strata >1,500 g (p = 0.003) but not different in the strata ≤1,500 g (78/93 [84%] vs. 72/90 [80%], p = 0.57). There was no effect when the experience level of residents exceeded 6 months. Neonatologists’ overall success rate and time to successful cannulation did not differ significantly in both weight strata. Conclusion: Transillumination improves the first-attempt success rate of peripheral vein cannulation performed by pediatric residents in neonates >1,500 g, while no benefit was found in infants ≤1,500 g.

Term and preterm neonates undergoing intensive care medicine frequently require peripherally inserted vein catheters (PIVCs) [1]. Multiple puncture attempts in this population is associated with pain exposure and potentially increased risk of complications such as hematoma and nerve damage, increased staff requirements, and in emergency scenarios, may delay life-saving treatment [2‒7]. As a consequence, the American Academy of Pediatrics states that the number of painful procedures in neonates should be reduced [8]. Effective cannulation of peripheral veins therefore represents a major goal in neonatal medicine. However, first-attempt success rates of PIVC insertion in clinical trials are non-satisfying and reported to be only 45–59% in neonate-specific study populations [9, 10]. To improve the PIVC placement success rate in children, different technical devices facilitating vein visualization were evaluated in several clinical studies [11‒19]. Primary success rates in these trials are equivocal with some studies demonstrating improvements, while others found no benefit or even lower success rates [11, 20]. The studies differ substantially in terms of included study subjects (neonatal, pediatric, or mixed populations), experience level of the operators, and technical devices used. These differences might have mitigated potential benefits of the intervention in the neonatal population. The primary goal of this randomized controlled trial (RCT) is to assess the effect of a red light-emitting diode (LED) transilluminator on the first-attempt PIVC insertion success rate compared to the standard procedure and pediatric residents as the primary operator in a weight-stratified neonatal study cohort.

Design and Study Population

This non-blinded single-center RCT was conducted in a German perinatal level III center between July 2018 and January 2021. The study is registered at the German Clinical Trial Registry (DRKS00015181) and was approved by the local Institutional Review Board prior to study start (AZ 61/18). Subjects were eligible if they met the following inclusion criteria: all preterm and term neonates admitted to the neonatal unit requiring nonemergency PIVC insertion. Infants were ineligible (or not randomized in cases of available informed consent) if emergency intravenous access was required, if hemodynamically unstable (defined as mean arterial blood pressure < gestational age [GA]) or considered too sick to participate as decided by the responsible neonatologist.

Study Procedure and Intervention

Infants meeting the inclusion criteria were randomly allocated to receive one of the two following interventions: in infants randomized to the transillumination group, a red LED-based transilluminator (ASTODIA®, Stihler Electronic GmbH, Stuttgart, Germany) was used to visualize the veins (shown in Fig. 1) with the surrounding light dimmed. In infants allocated to the control group, no device to visualize veins for PIVC insertion was permitted. The primary operator in each study group was a pediatric resident as they perform PIVC insertion routinely in Germany. Before participating in this trial, each pediatric resident attempted a minimum of six explanatory and supervised PIVC insertions (standard and transillumination, respectively). The transillumination device has been implemented for study purposes only and has not been routinely used in daily clinical practice. Within both study groups, the PIVC insertion followed a standardized protocol: a neonatal nurse was responsible for comforting the infant, and oral administration of 20% dextrose 1–2 min prior to the vein puncture was given. Time measurement was recorded by a neonatal nurse using a stopwatch and started when the pediatric resident was ready to start scanning for available veins. After effective skin disinfection (30 s), PIVC insertion was performed using either 24-gauge (BD Insyte® catheter, Becton Dickinson, UT, USA) or 26-gauge (Kliniject® catheter KLINIKA, Usingen, Germany) catheters as considered appropriate by the operator. Successful insertion of the PIVC was confirmed by flushing the intravenous catheter with normal saline, and timing was stopped when no visible extravasation occurred. After three nonsuccessful attempts by the pediatric resident, a neonatologist performed up to three further attempts using the initially allocated intervention. After the study procedure, the operator noted relevant demographic variables and outcome data on a hard-copy case report form.

Fig. 1.

LED-based transilluminator.

Fig. 1.

LED-based transilluminator.

Close modal

Randomization and Blinding

Study subjects were stratified by actual body weight at the time of randomization (≤1,500 g and >1,500 g). Randomization was single-sequence-based, and infants were allocated to the transillumination group or control group using sealed opaque envelopes generated by the research team. The participants were enrolled by the research team and the performers and randomly assigned to their group by the performers. Participating study subjects could be enrolled multiple times during their hospital stay if repeated PIVC insertion was necessary. Therefore, study subjects were allowed to switch weight strata with increasing body weight, and for every new required PIVC, patients were re-randomized. Crossover after allocation was not permitted. Due to the nature of the intervention, blinding of operators and participating staff members was not possible.

Study Outcomes

The primary outcome was successful PIVC insertion on first attempt performed by a pediatric resident. Secondary outcomes included success rate within two or three attempts by pediatric residents, success rate at first and within two or three attempts by neonatologists, and time to successful PIVC insertion at resident’s first attempt. Patient and operator demographic variables (GA at birth, GA at randomization, birthweight, weight at randomization, sex, ethnicity, catheter size, reason for PIVC insertion, experience level of the resident in months of residency training, and puncture site) were documented as well.

Sample Size

The success rate at first attempt in a previous study with a neonate-specific population was 45% [9]. Assuming that the LED transilluminator increases the success rate at first attempt from 50% to 65%, this would lead to a number needed to treat of 7 and was considered as a clinically important difference. Based on an alpha of 0.05 and a power of 80%, 170 patients per arm are needed to detect the 15% increase in first-attempt success. To account for potential dropouts, the sample size was increased by approximately 10% (190 procedures per study arm). Because body weight at the time of randomization may have a significant impact on the primary outcome, we a priori determined two different weight strata (strata 1 ≤1,500 g and strata 2 >1,500 g). The final calculated sample size for each weight strata was therefore 380 procedures.

Statistical Analysis

All statistical analyses were performed using IBM® SPSS® Statistics (Version 25, SPSS Inc., Chicago, IL, USA), the statistical software R (Version 4.1.1, 2021), and Microsoft® Excel® 2013 (Version 15.0.5423.1000, Redmond, WA, USA). For statistical comparisons of categorical variables (PIVC insertion success rate, sex, catheter size, reason for PIVC insertion, and puncture site), a two-sided Fisher’s exact test was applied, and the Mann-Whitney-U test was used to compare numerical variables (time to successful PIVC insertion at first attempt, GA at birth, GA at randomization, birthweight, weight at randomization, experience level of the resident in months). Data are presented as median and interquartile range (IQR) for numerical variables and percentage for categorical variables. For the primary and secondary outcomes, the unadjusted absolute increase in success rates between the two study groups was calculated and presented with 95% confidence interval based on Wald’s statistic. The number needed to treat to avoid failure at placement was calculated. A p value of <0.05 was considered statistically significant. A logistic regression model was calculated with success at first attempt as the primary outcome, taking the experience level of the resident or puncture site into account as a possible confounder. Post hoc Tukey tests with adjustment for multiple comparisons were performed if appropriate using the emmeans R package (Version 17.7.5, 2022). Time to successful PIVC insertion was analyzed using a Mann-Whitney U-test as data show a skewed distribution. All analyses were performed as per the intention-to-treat principle. An interim analysis after reaching a sample size of 50% within each weight strata was performed.

Study Population and Randomization

Between July 2018 and January 2021, custodians of 473 eligible neonates were approached for consent. We received consent for 405 (86%) neonates. A total of 127 patients were excluded as they had no need for a new PIVC insertion. Because study subjects could be re-enrolled, 570 PIVC insertions (190 ≤1,500 g, 380 >1,500 g) were performed in 278 neonates in this study. A total of 559 cases were included in the intention-to-treat analysis (183 ≤1,500 g; 376 >1,500 g), and 11 case report forms got lost during daily clinical routine (shown in Fig. 2). After the results of the interim analysis in March 2020, recruitment in the strata ≤1,500 g was stopped after randomization of 190 procedures due to a change of peripheral vein access policy in our unit and an insignificant difference regarding the primary outcome (47 in the transillumination vs. 49% in the control group). From this interim analysis, it seemed unlikely that ongoing recruitment would have affected the study results in a significant manner.

Fig. 2.

CONSORT flow diagram.

Fig. 2.

CONSORT flow diagram.

Close modal

Baseline Characteristics

Demographic variables and baseline characteristics are presented in Table 1. Except for the experience level of the residents in the strata ≤1,500 g and the primary puncture site in the strata >1,500 g, demographics were comparable in both groups. The median experience level in the strata ≤1,500 g was 20 months (IQR 12–30) in the transillumination group and 16 months (IQR 7–22) in the control group (p = 0.001). In logistic regression analysis, this difference did not affect the primary outcome (estimated log-odds change by using transillumination = 0.7975, p = 0.395).

Table 1.

Neonatal and pediatric residents’ baseline characteristics

Strata
≤1,500 g>1,500 g
overall (N = 183)transillumination group (N = 93)control group (N = 90)paoverall (N = 376)transillumination group (N = 188)control group (N = 188)pa
Neonatal baseline characteristics 
 GA at birth, median [IQR], weeks 28+0 [26+1–30+0] 27+3 [26+0–30+0] 28+4 [26+1–30+0] 0.40 36+3 [30+6–39+5] 36+0 [30+0–39+4] 37+5 [30+6–39+6] 0.22 
 GA at randomization, median [IQR], weeks 29+5 [28+3–31+3] 29+4 [28+2–31+3] 29+6 [28+4–31+2] 0.58 40+1 [36+2–41+6] 39+5 [36+3–41+4 ] 40+4 [36+0–42+1] 0.30 
 Age at randomization, median [IQR], days 5 [3–18] 5 [3–19] 6 [3–18] 0.78 12 [4–52] 12 [3–56] 12 [4–51] 0.85 
 Birthweight, median [IQR], g 900 [660–1,155] 900 [660–1,180] 905 [690–1,172] 0.60 2,650 [1,570–3,380] 2,340 [1,440–3,350] 2,774 [1,600–3,480] 0.21 
 Weight at randomization, median [IQR], g 1,090 [900–1,300] 1,080 [900–1,300] 1,100 [890–1,300] 0.82 3,170 [2,345–3,788] 3,135 [2,252–3,662] 3,215 [2,490–3,900] 0.23 
Sex    0.24    0.66 
 Male, n [%] 87 [48] 40 [43] 47 [52]  258 [69] 127 [68] 131 [70]  
 Female, n [%] 96 [52] 53 [57] 43 [48]  117 [31] 61 [32] 56 [30]  
 Diverse, n [%] 0 [0] 0 [0] 0 [0]  1 [0] 0 [0] 1 [1]  
Ethnicity    0.20    0.58 
 Afro-American, n [%] 7 [4] 2 [2] 5 [6]  7 [2] 3 [2] 4 [2]  
 Asian, n [%] 2 [1] 2 [2] 0 [0]  12 [3] 8 [4] 4 [2]  
 Caucasian, n [%] 174 [95] 89 [96] 85 [94]  357 [95] 177 [94] 180 [96]  
Reason for peripheral vein catheter insertion    0.70    0.78 
 Antibiotic therapy, n [%] 47 [26] 24 [26] 23 [26]  214 [57] 105 [56] 109 [58]  
 Parenteral nutrition, n [%] 84 [46] 39 [42] 45 [50]  104 [28] 55 [30] 49 [26]  
 Antibiotic therapy and parenteral nutrition, n [%] 36 [20] 19 [20] 17 [19]  22 [6] 10 [5] 12 [6]  
 Transfusion, n [%] 11 [6] 7 [8] 4 [4]  14 [4] 6 [3] 8 [4]  
 Sedation, n [%] 1 [1] 1 [1] 0 [0]  10 [3] 4 [2] 6 [3]  
 Antibiotic therapy and transfusion, n [%] 0 [0] 0 [0] 0 [0]  1 [0] 1 [1] 0 [0]  
 Antibiotic therapy and misc., n [%] 0 [0] 0 [0] 0 [0]  2 [1] 1 [1] 1 [1]  
 Misc. (antimycotics, drugs, contrast agents), n [%] 4 [2] 3 [3] 1 [1]  8 [2] 6 [3] 2 [1]  
 Missing data, n [%] 0 [0] 0 [0] 0 [0]  1 [0] 0 [0] 1 [1]  
Catheter size    0.55    0.38 
 24 gauge, n [%] 141 [77] 71 [76] 70 [78]  364 [97] 183 [97] 181 [96]  
 26 gauge, n [%] 30 [16] 14 [15] 16 [18]  9 [2] 5 [3] 4 [2]  
 26 and 24 gauge used within 3 attempts, n [%] 12 [7] 8 [9] 4 [4]  3 [1] 0 [0] 3 [2]  
Puncture site of primary puncture    0.06    0.02 
 Hand, n [%] 126 [69] 64 [69] 62 [69]  302 [80] 157 [84] 145 [77]  
 Foot, n [%] 38 [21] 15 [16] 23 [26]  59 [16] 24 [13] 35 [19]  
 Forearm, n [%] 14 [8] 11 [12] 3 [3]  6 [2] 4 [2] 2 [1]  
 Leg, n [%] 4 [2] 3 [3] 1 [1]  2 [1] 2 [1] 0 [0]  
 Head, n [%] 0 [0] 0 [0] 0 [0]  5 [1] 0 [0] 5 [3]  
 Missing data, n [%] 1 [1] 0 [0] 1 [1]  2 [1] 1 [1] 1 [1]  
Pediatric residents’ baseline characteristics 
 Experience level of the resident, median [IQR], months 18 [8–27] 20 [12–30] 16 [7–22] 0.001 12 [5–20] 14 [5–24] 12 [5–18] 0.38 
Strata
≤1,500 g>1,500 g
overall (N = 183)transillumination group (N = 93)control group (N = 90)paoverall (N = 376)transillumination group (N = 188)control group (N = 188)pa
Neonatal baseline characteristics 
 GA at birth, median [IQR], weeks 28+0 [26+1–30+0] 27+3 [26+0–30+0] 28+4 [26+1–30+0] 0.40 36+3 [30+6–39+5] 36+0 [30+0–39+4] 37+5 [30+6–39+6] 0.22 
 GA at randomization, median [IQR], weeks 29+5 [28+3–31+3] 29+4 [28+2–31+3] 29+6 [28+4–31+2] 0.58 40+1 [36+2–41+6] 39+5 [36+3–41+4 ] 40+4 [36+0–42+1] 0.30 
 Age at randomization, median [IQR], days 5 [3–18] 5 [3–19] 6 [3–18] 0.78 12 [4–52] 12 [3–56] 12 [4–51] 0.85 
 Birthweight, median [IQR], g 900 [660–1,155] 900 [660–1,180] 905 [690–1,172] 0.60 2,650 [1,570–3,380] 2,340 [1,440–3,350] 2,774 [1,600–3,480] 0.21 
 Weight at randomization, median [IQR], g 1,090 [900–1,300] 1,080 [900–1,300] 1,100 [890–1,300] 0.82 3,170 [2,345–3,788] 3,135 [2,252–3,662] 3,215 [2,490–3,900] 0.23 
Sex    0.24    0.66 
 Male, n [%] 87 [48] 40 [43] 47 [52]  258 [69] 127 [68] 131 [70]  
 Female, n [%] 96 [52] 53 [57] 43 [48]  117 [31] 61 [32] 56 [30]  
 Diverse, n [%] 0 [0] 0 [0] 0 [0]  1 [0] 0 [0] 1 [1]  
Ethnicity    0.20    0.58 
 Afro-American, n [%] 7 [4] 2 [2] 5 [6]  7 [2] 3 [2] 4 [2]  
 Asian, n [%] 2 [1] 2 [2] 0 [0]  12 [3] 8 [4] 4 [2]  
 Caucasian, n [%] 174 [95] 89 [96] 85 [94]  357 [95] 177 [94] 180 [96]  
Reason for peripheral vein catheter insertion    0.70    0.78 
 Antibiotic therapy, n [%] 47 [26] 24 [26] 23 [26]  214 [57] 105 [56] 109 [58]  
 Parenteral nutrition, n [%] 84 [46] 39 [42] 45 [50]  104 [28] 55 [30] 49 [26]  
 Antibiotic therapy and parenteral nutrition, n [%] 36 [20] 19 [20] 17 [19]  22 [6] 10 [5] 12 [6]  
 Transfusion, n [%] 11 [6] 7 [8] 4 [4]  14 [4] 6 [3] 8 [4]  
 Sedation, n [%] 1 [1] 1 [1] 0 [0]  10 [3] 4 [2] 6 [3]  
 Antibiotic therapy and transfusion, n [%] 0 [0] 0 [0] 0 [0]  1 [0] 1 [1] 0 [0]  
 Antibiotic therapy and misc., n [%] 0 [0] 0 [0] 0 [0]  2 [1] 1 [1] 1 [1]  
 Misc. (antimycotics, drugs, contrast agents), n [%] 4 [2] 3 [3] 1 [1]  8 [2] 6 [3] 2 [1]  
 Missing data, n [%] 0 [0] 0 [0] 0 [0]  1 [0] 0 [0] 1 [1]  
Catheter size    0.55    0.38 
 24 gauge, n [%] 141 [77] 71 [76] 70 [78]  364 [97] 183 [97] 181 [96]  
 26 gauge, n [%] 30 [16] 14 [15] 16 [18]  9 [2] 5 [3] 4 [2]  
 26 and 24 gauge used within 3 attempts, n [%] 12 [7] 8 [9] 4 [4]  3 [1] 0 [0] 3 [2]  
Puncture site of primary puncture    0.06    0.02 
 Hand, n [%] 126 [69] 64 [69] 62 [69]  302 [80] 157 [84] 145 [77]  
 Foot, n [%] 38 [21] 15 [16] 23 [26]  59 [16] 24 [13] 35 [19]  
 Forearm, n [%] 14 [8] 11 [12] 3 [3]  6 [2] 4 [2] 2 [1]  
 Leg, n [%] 4 [2] 3 [3] 1 [1]  2 [1] 2 [1] 0 [0]  
 Head, n [%] 0 [0] 0 [0] 0 [0]  5 [1] 0 [0] 5 [3]  
 Missing data, n [%] 1 [1] 0 [0] 1 [1]  2 [1] 1 [1] 1 [1]  
Pediatric residents’ baseline characteristics 
 Experience level of the resident, median [IQR], months 18 [8–27] 20 [12–30] 16 [7–22] 0.001 12 [5–20] 14 [5–24] 12 [5–18] 0.38 

IQR, interquartile range.

aFisher’s exact test or Mann-Whitney U test as appropriate.

PIVC Insertion Success Rate and Time to Successful PIVC Insertion

Neonates ≤1,500 g

The first-attempt success rate in neonates ≤1,500 g performed by pediatric residents was not significantly different between the transillumination and the control group (47 [44/93] vs. 49% [44/90], p = 0.88, Table 2). The success rates within two and three attempts by pediatric residents were also comparable (Table 2). When the procedure was performed by a neonatologist, there were no significant differences between both groups in the success rate (Table 2). Time to successful PIVC insertion at resident’s first attempt was comparable between the groups (Table 3).

Table 2.

Success rate in % (ratio/total) and absolute increase in success rates with a 95% confidence interval in both groups

Transillumination group, % [ratio/total]Control group, % [ratio/total]Absolute increase in success rate percentage point [95% confidence interval]pa
≤1,500 g 
 Success rate at first attempt by pediatric residents 47 [44/93] 49 [44/90] −2 [−16 to 13] 0.88 
 Success rate within two attempts by pediatric residents 69 [64/93] 70 [63/90] −1 [−15 to 12] 0.87 
 Success rate within three attempts by pediatric residents 84 [78/93] 80 [72/90] 4 [−7 to 15] 0.57 
 Success rate at first attempt by neonatologists 60 [9/15] 72 [13/18] −12 [−45 to 20] 0.49 
 Success rate within two attempts by neonatologists 87 [13/15] 78 [14/18] 9 [−17 to 35] 0.66 
 Success rate within three attempts by neonatologists 93 [14/15] 78 [14/18] 16 [−7 to 39] 0.35 
>1,500 g 
 Success rate at first attempt by pediatric residents 55 [103/188] 34 [64/188] 21 [11–31] <0.001 
 Success rate within two attempts by pediatric residents 78 [146/188] 63 [118/188] 15 [6–24] 0.002 
 Success rate within three attempts by pediatric residents 86 [162/188] 73 [138/188] 13 [5–21] 0.003 
 Success rate at first attempt by neonatologists 58 [15/26] 48 [24/50] 10 [−14 to 33] 0.48 
 Success rate within two attempts by neonatologists 73 [19/26] 68 [34/50] 5 [−16 to 26] 0.79 
 Success rate within three attempts by neonatologists 88 [23/26] 76 [38/50] 12 [−5 to 30] 0.24 
Transillumination group, % [ratio/total]Control group, % [ratio/total]Absolute increase in success rate percentage point [95% confidence interval]pa
≤1,500 g 
 Success rate at first attempt by pediatric residents 47 [44/93] 49 [44/90] −2 [−16 to 13] 0.88 
 Success rate within two attempts by pediatric residents 69 [64/93] 70 [63/90] −1 [−15 to 12] 0.87 
 Success rate within three attempts by pediatric residents 84 [78/93] 80 [72/90] 4 [−7 to 15] 0.57 
 Success rate at first attempt by neonatologists 60 [9/15] 72 [13/18] −12 [−45 to 20] 0.49 
 Success rate within two attempts by neonatologists 87 [13/15] 78 [14/18] 9 [−17 to 35] 0.66 
 Success rate within three attempts by neonatologists 93 [14/15] 78 [14/18] 16 [−7 to 39] 0.35 
>1,500 g 
 Success rate at first attempt by pediatric residents 55 [103/188] 34 [64/188] 21 [11–31] <0.001 
 Success rate within two attempts by pediatric residents 78 [146/188] 63 [118/188] 15 [6–24] 0.002 
 Success rate within three attempts by pediatric residents 86 [162/188] 73 [138/188] 13 [5–21] 0.003 
 Success rate at first attempt by neonatologists 58 [15/26] 48 [24/50] 10 [−14 to 33] 0.48 
 Success rate within two attempts by neonatologists 73 [19/26] 68 [34/50] 5 [−16 to 26] 0.79 
 Success rate within three attempts by neonatologists 88 [23/26] 76 [38/50] 12 [−5 to 30] 0.24 

aFisher’s exact test.

Table 3.

Time to successful peripheral vein catheter insertion at resident’s first attempt in median (IQR) in both groups

StrataTransillumination groupControl grouppa
median [IQR]Nmedian [IQR]N
≤1,500 g 174 s [120–252] 44 172 s [120–270] 43 0.94 
>1,500 g 180 s [114–278] 103 161 s [100–244] 64 0.28 
StrataTransillumination groupControl grouppa
median [IQR]Nmedian [IQR]N
≤1,500 g 174 s [120–252] 44 172 s [120–270] 43 0.94 
>1,500 g 180 s [114–278] 103 161 s [100–244] 64 0.28 

One successful procedure in the transillumination group was not timed.

IQR, interquartile range.

aMann-Whitney U test.

Neonates >1,500 g

When an LED transilluminator was used, the first-attempt success rate by pediatric residents was significantly higher at 55% (103/188) compared to the standard procedure at 34% (64/188) (number needed to treat: 5, p < 0.001, Table 2). The success rate within two and three attempts by pediatric residents was also significantly higher in the transillumination group. Neonatologists’ success rate was not significantly affected by the study intervention (Table 2). Time to successful PIVC insertion at resident’s first attempt was comparable between the groups (Table 3).

Success Rate in Relation to the Clinical Level of Experience

The LED transilluminator significantly improved the success rate at first and within three attempts for pediatric residents with experience of up to 6 months. This effect became less pronounced with increasing clinical experience. There was no significantly improvement when the experience level of residents exceeded 6 months (Table 4).

Table 4.

Success rate in % (ratio/total), absolute increase in success rates with a 95% confidence interval, and the number needed to treat with a 95% confidence interval according to the level of experience across both weight strata

Transillumination group, % [ratio/total]Control group, % [ratio/total]Absolute increase in success rate percentage point [95% confidence interval]Number needed to treat [95% confidence interval]pa
<6 months 
 Success rate at first attempt 56 [44/79] 28 [24/87] 28 [14–43] 4 [3–8] <0.001 
 Success rate within two attempts 77 [61/79] 52 [45/87] 26 [12–40] 4 [3–9] 0.001 
 Success rate within three attempts 85 [67/79] 66 [57/87] 19 [7–32] 6 [4–16] 0.005 
6–12 months 
 Success rate at first attempt 49 [18/37] 34 [20/58] 14 [−6 to 34] 8 [−17 to 3] 0.20 
 Success rate within two attempts 70 [26/37] 66 [38/58] 5 [−14 to 24] 22 [−7 to 5] 0.66 
 Success rate within three attempts 78 [29/37] 76 [44/58] 3 [−15 to 20] 40 [−7 to 6] 0.81 
12–18 months 
 Success rate at first attempt 52 [30/58] 50 [29/58] 2 [−17 to 20] 58 [−7 to 6] 
 Success rate within two attempts 78 [45/58] 79 [46/58] −2 [−17 to 13] −58 [−6 to 8] 
 Success rate within three attempts 86 [50/58] 88 [51/58] −2 [−14 to 10] −58 [−8 to 10] 
18–24 months 
 Success rate at first attempt 55 [26/47] 54 [15/28] 2 [−22 to 25] 58 [−5 to 4] 
 Success rate within two attempts 81 [38/47] 82 [23/28] −1 [−19 to 17] −78 [−6 to 6] 
 Success rate within three attempts 89 [42/47] 82 [23/28] 7 [−10 to 24] 14 [−11 to 5] 0.49 
>24 months 
 Success rate at first attempt 48 [29/60] 43 [20/47] 6 [−13 to 25] 18 [−8 to 5] 0.56 
 Success rate within two attempts 67 [40/60] 62 [29/47] 5 [−13 to 23] 21 [−8 to 5] 0.68 
 Success rate within three attempts 87 [52/60] 74 [35/47] 12 [−3 to 27] 9 [−34 to 4] 0.14 
Transillumination group, % [ratio/total]Control group, % [ratio/total]Absolute increase in success rate percentage point [95% confidence interval]Number needed to treat [95% confidence interval]pa
<6 months 
 Success rate at first attempt 56 [44/79] 28 [24/87] 28 [14–43] 4 [3–8] <0.001 
 Success rate within two attempts 77 [61/79] 52 [45/87] 26 [12–40] 4 [3–9] 0.001 
 Success rate within three attempts 85 [67/79] 66 [57/87] 19 [7–32] 6 [4–16] 0.005 
6–12 months 
 Success rate at first attempt 49 [18/37] 34 [20/58] 14 [−6 to 34] 8 [−17 to 3] 0.20 
 Success rate within two attempts 70 [26/37] 66 [38/58] 5 [−14 to 24] 22 [−7 to 5] 0.66 
 Success rate within three attempts 78 [29/37] 76 [44/58] 3 [−15 to 20] 40 [−7 to 6] 0.81 
12–18 months 
 Success rate at first attempt 52 [30/58] 50 [29/58] 2 [−17 to 20] 58 [−7 to 6] 
 Success rate within two attempts 78 [45/58] 79 [46/58] −2 [−17 to 13] −58 [−6 to 8] 
 Success rate within three attempts 86 [50/58] 88 [51/58] −2 [−14 to 10] −58 [−8 to 10] 
18–24 months 
 Success rate at first attempt 55 [26/47] 54 [15/28] 2 [−22 to 25] 58 [−5 to 4] 
 Success rate within two attempts 81 [38/47] 82 [23/28] −1 [−19 to 17] −78 [−6 to 6] 
 Success rate within three attempts 89 [42/47] 82 [23/28] 7 [−10 to 24] 14 [−11 to 5] 0.49 
>24 months 
 Success rate at first attempt 48 [29/60] 43 [20/47] 6 [−13 to 25] 18 [−8 to 5] 0.56 
 Success rate within two attempts 67 [40/60] 62 [29/47] 5 [−13 to 23] 21 [−8 to 5] 0.68 
 Success rate within three attempts 87 [52/60] 74 [35/47] 12 [−3 to 27] 9 [−34 to 4] 0.14 

aFisher’s exact test.

This RCT assessed the effect of an LED transilluminator on the first-attempt PIVC insertion success rate in a neonatal study population that was further segregated into two weight strata (≤1,500 g and >1,500 g, respectively) and different levels of experience of pediatric residents as the primary operators. We found a significant improvement in first attempt and overall success rate in the strata >1,500 g when the LED transilluminator was used by pediatric residents.

Of interest, the primary success rate of pediatric residents in the strata ≤1,500 g did not differ significantly between the study groups. This is in line with the findings by Phipps et al. where benefits of vein visualization were most clearly seen in more mature neonates [10]. The most obvious explanation for the improved success rate in more mature neonates could be the higher amount of subcutaneous fat, a condition where transillumination most effectively visualizes subcutaneous veins. Another possible explanation is the more difficult handling of the transilluminator in infants ≤1,500 g due to the disproportionally smaller size of the limbs compared to the size of the transilluminator.

Skills of the operator have been identified as one key determinant for successful venous puncture [2]. Our data support this finding and demonstrate that the use of the transilluminator can significantly improve success rates for less experienced staff (Table 4).

Phipps et al. [10] assessed a near-infrared-based vein visualization device for the placement of peripheral inserted central venous catheters and found no difference on the success at first attempt but an improved overall success rate within three attempts. Comparison of their outcome data with our study results is difficult. First, the device used in their trial uses near-infrared technology that directly visualizes hemoglobin parts of the flowing or non-flowing blood inside or outside the superficial veins, while the transillumination technique rather directly visualizes the anatomical vein structure and may more effectively allow the differentiation of the vein from the surrounding transilluminated tissue. Second, the primary operator in their trial were members of an expert neonatal nurse practitioners/registered nurses PICC-line team, and although no particular experience level was reported, we presume a higher level of operator experience compared to the performing residents in our trial [10].

The success rates of the primary operator without vein visualization in our study are below those reported in other studies [21]. The most likely explanation is the lower level of experience of the pediatric residents (Table 4).

Strengths of our study include an adequate sample size that allows confirmatory interpretation of our study results, particularly in the weight strata >1,500 g. The clear definition of the primary operator is another strength of this trial, and we were able to collect important performance data on the success rate of pediatric residents. The primary success rate of pediatric residents (representative for all less experienced staff members and trainees) in the control group of only 34% requires action and should prompt supervisors to seek for improved teaching strategies, including the use of skin transillumination. This strategy may further help to improve first-attempt success and, as a consequence, reduce procedure-related pain in this vulnerable population [7, 8]. However, our study also has limitations. Due to the nature of the intervention, blinding within this trial was not possible. Earlier termination in the low-weight strata due to a change in our vein access policy was mainly based on the observation of very short longevity of peripheral vein catheters in infants ≤1,500 g and the consecutive high PIVC insertion exposure. We therefore shifted toward the use of peripheral inserted central vein catheters in this study population. However, since there were no differences after 190 randomized procedures, we believed that it was unlikely that ongoing recruitment would have affected the study results in a significant manner.

In conclusion, the use of a red LED-based transilluminator significantly increased the success rate of first attempts of PIVC insertion in the neonatal population with an actual weight >1,500 g for pediatric residents, and this effect was most pronounced for unexperienced pediatric residents. Vein visualization therefore is a very useful tool to increase success rates of PIVC insertion and to reduce the overall pain burden and the risk for other procedure-related complications associated with PIVC insertion in this vulnerable population.

We thank the staff of the Neonatal Ward of the Center for Pediatrics and Youth Medicine in Giessen for their support in collecting the data and the parents who agreed that their neonates could participate in our study. We further thank Dr. Lars Mense (Department of Pediatrics, Divison of Neonatology and Pediatric Intensive Care Medicine, TU Dresden, Medical Faculty Carl Gustav Carus, Dresden, Germany) who carefully reviewed the manuscript with his expertise and improved the manuscript with valuable suggestions.

This study protocol was reviewed and approved by the Medical Ethics Committee of the Justus-Liebig-University of Giessen, approval number AZ 61/18. Written informed consent was obtained from the parents of the participants to participate in the study.

The other authors have no conflicts of interest to disclose.

No funding was secured for this study.

Samantha Hinterstein: methodology, investigation, data curation, formal analysis, writing – drafting the initial manuscript, and writing – review or editing of the manuscript. Prof Harald Ehrhardt: conceptualization/design, investigation, and writing – review or editing of the manuscript. Prof Klaus-Peter Zimmer: conceptualization/design, supervision/oversight, resources, and writing – review or editing of the manuscript. Dr. Anita Windhorst: methodology, formal analysis, and writing – review or editing of the manuscript. Dr. Judith Kappesser: conceptualization/design and writing – drafting the initial manuscript. Prof Christiane Hermann: conceptualization/design, methodology, and writing – review or editing of the manuscript. Dr. Rahel Schuler and Dr. Markus Waitz: conceptualization/design, methodology, investigation, supervision/oversight, formal analysis, writing – drafting the initial manuscript, and writing – review or editing of the manuscript. All the authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Additional Information

Rahel Schuler and Markus Waitz contributed equally as senior authors.Clinical Trial Registration: German Clinical Trials Register, DRKS00015181, www.drks.de, date of registration: 07/31/2018.

The data that support the findings of this study are not publicly available due to containing information that could compromise the privacy of research participants but are available from the corresponding author (Rahel Schuler, Rahel.Schuler@paediat.med.uni-giessen.de) upon reasonable request.

1.
Davis
MBH
,
Takashima
M
,
Girgenti
C
,
Ullman
AJ
.
An international survey of pediatric and neonatal clinicians’ vascular access practice: PediSIG assessment of vascular access, education, and support (PAVES) catheter selection
.
Br J Nurs
.
2020
;
29
(
14
):
S40
8
.
2.
Jöhr
M
,
Berger
TM
.
Venous access in children: state of the art
.
Curr Opin Anaesthesiol
.
2015
;
28
(
3
):
314
20
.
3.
Unbeck
M
,
Förberg
U
,
Ygge
B-M
,
Ehrenberg
A
,
Petzold
M
,
Johansson
E
.
Peripheral venous catheter related complications are common among paediatric and neonatal patients
.
Acta Paediatr
.
2015
;
104
(
6
):
566
74
.
4.
Ramasethu
J
.
Complications of vascular catheters in the neonatal intensive care unit
.
Clin Perinatol
.
2008
;
35
(
1
):
199
222
.
5.
Hermansen
MC
,
Hermansen
MG
.
Intravascular catheter complications in the neonatal intensive care unit
.
Clin Perinatol
.
2005
;
32
(
1
):
141
56
.
6.
Liossis
G
,
Bardin
C
,
Papageorgiou
A
.
Comparison of risks from percutaneous central venous catheters and peripheral lines in infants of extremely low birth weight: a cohort controlled study of infants <1,000 g
.
J Matern Fetal Neonatal Med
.
2003
;
13
(
3
):
171
4
.
7.
Taddio
A
,
Shah
V
,
Gilbert-MacLeod
C
,
Katz
J
.
Conditioning and hyperalgesia in newborns exposed to repeated heel lances
.
JAMA
.
2002
;
288
(
7
):
857
61
.
8.
American Academy of Pediatrics Committee on Fetus and Newborn
;
American Academy of Pediatrics Section on Surgery, Canadian Paediatric Society Fetus and Newborn Committee
;
Batton
DG
,
Barrington
KJ
,
Wallman
C
.
Prevention and management of pain in the neonate: an update
.
Pediatrics
.
2006
;
118
(
5
):
2231
41
.
9.
Legemaat
M
,
Carr
PJ
,
van Rens
RM
,
van Dijk
M
,
Poslawsky
IE
,
van den Hoogen
A
.
Peripheral intravenous cannulation: complication rates in the neonatal population: a multicenter observational study
.
J Vasc Access
.
2016
;
17
(
4
):
360
5
.
10.
Phipps
K
,
Modic
A
,
O’Riordan
MA
,
Walsh
M
.
A randomized trial of the Vein Viewer versus standard technique for placement of peripherally inserted central catheters (PICCs) in neonates
.
J Perinatol
.
2012
;
32
(
7
):
498
501
.
11.
Vyas
V
,
Sharma
A
,
Goyal
S
,
Kothari
N
.
Infrared vein visualization devices for ease of intravenous access in children: hope versus hype
.
Anaesthesiol Intensive Ther
.
2021
;
53
(
1
):
69
78
.
12.
Elkhunovich
M
,
Barreras
J
,
Bock Pinero
V
,
Ziv
N
,
Vaiyani
A
,
Mailhot
T
.
The use of ultrasound for peripheral IV placement by vascular access team nurses at a tertiary children’s hospital
.
J Vasc Access
.
2017
;
18
(
1
):
57
63
.
13.
Benkhadra
M
,
Collignon
M
,
Fournel
I
,
Oeuvrard
C
,
Rollin
P
,
Perrin
M
, et al
.
Ultrasound guidance allows faster peripheral IV cannulation in children under 3 years of age with difficult venous access: a prospective randomized study
.
Paediatr Anaesth
.
2012
;
22
(
5
):
449
54
.
14.
Curtis
SJ
,
Craig
WR
,
Logue
E
,
Vandermeer
B
,
Hanson
A
,
Klassen
T
.
Ultrasound or near-infrared vascular imaging to guide peripheral intravenous catheterization in children: a pragmatic randomized controlled trial
.
CMAJ (Can Med Assoc J)
.
2015
;
187
(
8
):
563
70
.
15.
Doniger
SJ
,
Ishimine
P
,
Fox
JC
,
Kanegaye
JT
.
Randomized controlled trial of ultrasound-guided peripheral intravenous catheter placement versus traditional techniques in difficult-access pediatric patients
.
Pediatr Emerg Care
.
2009
;
25
(
3
):
154
9
.
16.
Hosokawa
K
,
Kato
H
,
Kishi
C
,
Kato
Y
,
Shime
N
.
Transillumination by light-emitting diode facilitates peripheral venous cannulations in infants and small children
.
Acta Anaesthesiol Scand
.
2010
;
54
(
8
):
957
61
.
17.
Goren
A
,
Laufer
J
,
Yativ
N
,
Kuint
J
,
Ben Ackon
M
,
Rubinshtein
M
, et al
.
Transillumination of the palm for venipuncture in infants
.
Pediatr Emerg Care
.
2001
;
17
(
2
):
130
1
.
18.
Atalay
H
,
Erbay
H
,
Tomatir
E
,
Serin
S
,
Oner
O
.
The use of transillumination for peripheral venous access in paediatric anaesthesia
.
Eur J Anaesthesiol
.
2005
;
22
(
4
):
317
8
.
19.
Katsogridakis
YL
,
Seshadri
R
,
Sullivan
C
,
Waltzman
ML
.
Veinlite transillumination in the pediatric emergency department: a therapeutic interventional trial
.
Pediatr Emerg Care
.
2008
;
24
(
2
):
83
8
.
20.
Inal
S
,
Demir
D
.
Impact of peripheral venous catheter placement with vein visualization device support on success rate and pain levels in pediatric patients aged 0 to 3 years
.
Pediatr Emerg Care
.
2021
;
37
(
3
):
138
44
.
21.
Monasor-Ortolá
D
,
Cortés-Castell
E
,
Martínez-Pascual
C
,
Esteve-Ríos
A
,
Rizo-Baeza
MM
.
Factors influencing the success of peripheral venous access in neonates
.
J Pediatr Nurs
.
2019
;
47
:
e30
5
.