Indian Pediatrics 2004; 41:250-254
Use of Magnifying Lens to Aid Neonatal Umbilical Arterial Catheter Insertion
L. McGlone, L. Devlin, *D.A. Doherty and S. Patole
From the Department of Neonatal Pediatrics, King Edward Memorial Hospital for Women, Bagot Rad, Subiaco, Western Australia 6008, and *Women’s and Infant’s Research Foundation at King Edward Memorial Hospital for Women and School of Women’s and Infant’s Health at University of Western Australia.
Umbilical arterial catheters (UACs) are frequently used in neonatal intensive care units (NICU) for continuous monitoring of blood gases and arterial blood pressure(1-3). The conventional method of UAC insertion can be technically challenging and time- consuming at a time when rapid stabilization of the neonate is desired(4-5). Most of the difficulties in the insertion of a UAC in preterm neonates are due to the narrow vessel diameter. We therefore conducted a clinical trial to test the hypothesis that the use of a magnifying lens will facilitate insertion of UACs by making the vessel easier to identify and manipulate. Our aim was to evaluate the efficacy of a magnifying lens in reducing the time to insertion of UAC (primary outcome) and the frequency of successful insertion (secondary outcome).
Patients and Methods
This study was designed as a randomized controlled trial in a tertiary neonatal intensive care unit. Ethics committee approval was obtained prior to commencement of the study.
All neonates <33 weeks’ gestation requiring UAC for hemodynamic monitoring were eligible for enrolment in the study after informed parental consent was obtained. Neonates were excluded if there was presence of congenital malformations (e.g., diaphrag-matic hernia, exomphalos), known chromoso-mal aberrations, single umbilical artery, umbilical infection, or necrotising entero-colitis.
In the conventional method of UAC insertion (without the use of magnifying lens) all UACs were inserted under aseptic conditions with the neonate under a radiant warmer as per our unit protocol. Oxygen saturation and heart rate was monitored continuously throughout the procedure. Length of catheter insertion was determined by measuring the shoulder to umbilicus distance and referral to the graph(6). A size 3.7 or 5.0 French Guage UAC was used depending on the size of the neonate. All catheters were primed with normal saline prior to insertion. After cleaning the skin with chlorhexidine, the cord was transected parallel to the abdomen leaving a 1 cm umbilical stump. The cord stump was stabilized with artery forceps and the artery gently probed with the tip of curved iris forceps. The forceps were allowed to spring open, dilating the artery. The catheter tip was then introduced into the artery and advanced with steady pressure to the predetermined position. Confirmation of catheter placement was made by aspiration of free flow of blood followed by a radiograph. The assisting midwife timed catheter insertion using a LCD stopwatch. The watch was started when the cord was incised and stopped when either a pulsatile column of blood was aspirated into the UAC or the procedure abandoned due to failure of arterial cannulation.
The experimental method of UAC insertion was identical to the conventional method of UAC insertion except for the additional use of a commercially available power 2.0 × magnifying lens (Megaview- Easifit) mounted on a head mount loupe. The mount loupe was a headband, which allows the operator to manipulate the lens position. The use of this device ensured sterility of the operating field.
Sample size calculations were based on the previously published data(7) and our own pilot data generated by monitoring UAC insertion (n = 5 each) by two registrars with comparable experience on 10 neonates <33 weeks gestation indicating a 10% rate of failed insertions and a mean ± SD time of insertion as 310 ± 132.7 sec. Group sample sizes of 22 each were estimated to achieve 82% power to detect a 40% reduction in the insertion time (considered clinically significant) from the 330 sec (without the use of magnifying lens) to 200 seconds while using a magnifying lens (common SD =144).
A two-sided Mann-Whitney test or t-test was planned to be as appropriate depending on the data normality, at a 0.05 significance level.
Neonates eligible for enrolment were randomized into the study (stratifying for birth weight <1000 g and >1000 g) using a computer generated sequence of treatment allocations (conventional or experimental method of UAC insertion) stored in sealed opaque envelopes that were opened after the parental consent was obtained. Two separate random sequences were generated to allow for the stratification by birth weight. Four registrars with comparable experience in level III neonatal care participated in the study.
The primary endpoint of insertion time was analyzed using Mann-Whitney test or t-test, and the secondary endpoint of success of UAC insertion was compared using Fisher exact test.
A total of 44 neonates (conventional method = 23, experimental method = 21) completed the study (Table 1). Overall 16 (36%) neonates were born with birth weight <1000 g conventional: Nine (39%), experi-mental: 7 (33%). Groups were comparable with respect to gestational age at delivery, birth weight, intrauterine growth retardation, and CRIB scores.
Table I Characteristics of Infants Requiring UAC*
*value expressed as medians and interquartile ranges (IQR = Q25 - Q75). *SGA = small for gestational age ( <10th centile).
Table II Time and Success of UAC* Insertion
*Value expressed as medians and interquartile ranges (IQR = Q25 - Q75).
No significant differences were found between the median times for successful UAC insertion between the two study groups. (conventional: 88 sec (IQR 50 - 192, range 24-780), Experimental: 70 sec (IQR 55-222, range 27-642), P = 0.734.
Similarly, there were no significant differences between the two groups (conventional versus experimental method of UAC insertion) with respect to the frequency of successful UAC insertion. (Conventional: 19/23 (83%) vs Experimental: 13/21 (63%), P = 0.179). The median insertion times in the unsuccessful cases were 345 sec (IQR 205-529, range 163-900), and 900 sec (IQR 370-900, range 148-900) with no statistically significant differences (P = 0.683). The comparisons in this case however were based on small sample sizes (n = 4 and 8).
Group comparison performed at the completion of the study indicated that the median (and mean) times to UAC insertion were considerably shorter than anticipated before the study commenced (189.6 vs 167 seconds). This resulted in only 6% power to detect statistically significant differences between the study arms. Approximate sample sizes of 100 neonates per group would be required to detect a 40% reduction in the insertion times between median times of 187 sec and 112 sec with 80% power at 5% significance level using a sequential design with a single interim analysis. Given that the insertion times achieved in the two groups during the trial were very close and significantly shorter than those before the study, continuation of the study was considered unfeasible due to the very low probability of finding statistically significant differences between the study groups.
Insertion of the UAC plays a vital role in the stabilization and management of ill preterm neonates. Reducing the time to insert these catheters is desirable to minimize handling and allow rapid and appropriate interventions for the altered cardiorespiratory status at a critical time(4,5). Most of technical difficulties in the insertion of UACs are related to the small calibre of the vessel. Vasospasm, vascular perforation, subintimal cannulation and tunica intima invagination have also been proposed as factors respon-sible for failure of UAC cannulation (3,8).
Mehods to overcome the technical difficulties in insertion of UACs have received little attention in the past. Results of a randomized controlled trial by Squire et al.(7) have shown that compared with the conventional method the side-entry method led to successful UAC insertions more frequently and the time to successful insertions was significantly shorter. (Correct placement (n/N): side entry (15/16) vs conventional (11/16), P = 0.09), (time) to insertion in seconds: side entry (143 ± 24) vs conventional (330 ± 36), P = 0.01).
Our results (comparable but overall signi-ficantly reduced insertion times compared with our pilot data) may be explained by a great variability between the operators which was probably far more important than the benefits of using a magnifying glass to aid the UAC insertion. This was also important, as one of the two registrars whose UAC insertion timings were used for the pilot data could not participate in the clinical trial due to un-expected circumstances. Though considered comparable (with other three operators) before the trial, the UAC insertion skills of the replacement registrar may also have differed significantly during the trial. The other possibility could be an improved awareness (benefits of inclusion in a clinical trial) with possibly an improved performance in UAC insertion during the study. The side entry method described by Squire et al.(7) involves visual identification of the umbilical arteries through the cord stump, transection of the Wharton’s jelly to the artery, followed by incision only through the upper arterial wall to expose the lumen. We have found complete transection of the umbilical artery to be a significant problem with this method of UAC insertion especially in extremely low birth weight neonates (ELBW). Our median times and frequency of successful UAC insertions are significantly better than those reported by Squire et al.(7). The differences again could be explained by variability in operator skills. In summary given the importance of speedy and successful vascular access in stabilization of ELBW neonates it is important to study methods to optimize the success of con-ventional method of UAC insertion. Our results emphasize the importance of consider-ing the variability of (inter and intra) operator skills during the study before undertaking any such trials.
We are thankful to Dr. S. Rao for participation in the study and for obtaining some of the parental consents.
Contributors: LM was responsible for the first and final draft, and patient enrolment, LD contributed to the first draft, DAD was responsible for study design, statistical analysis, and interpretation. SP was involved in the concept, design, and supervision of the final draft.
Competing interests: None stated.