Evaluation of different needle gauge blood collection sets (23G/25G) in aged patients

Introduction

Laboratory blood examination is one of the most common diagnostic procedures in clinical medicine. The common way to obtain blood is via venous access with a needle connected to a blood collection set, often by single-use plastic tubes with straight or butterfly needles [1]. Although blood collection with a straight needle is gold standard due to lower risk for hemolysis and better sample quality, the introduction of butterfly needles led to widespread adoption in pediatrics and geriatrics due to greater convenience (better visibility of whether a blood vessel has been struck and the greater freedom of movement of the blood collection set during blood draw) [1].

Difficulties in blood draw can be challenging due to difficult vein condition or non-compliance with blood collection procedure. This is frequently challenging especially in aged patients due to thin or aged veins and frequent cognitive limitations [2]. In addition, difficulties in blood draws quite often result in underfilling of blood sampling tubes. Underfilling the collection tubes increases the risk of hemolysis, which consequently elevates activity levels of lactate dehydrogenase (LD) [3], [4], [5], [6], and hemolysis is one of the major preanalytical errors in laboratory medicine [7]. Since LD is used for diagnostic as well as prognostic monitoring in aged patients [8], 9], this may lead to massive clinical misinterpretations [4].

To ensure a successful and painless blood collection, the needle should not be moved within the vein during blood draw. Especially in cases of multiple tube collection, this can be challenging. Furthermore, at the end of the blood sampling procedure, the tourniquet must be released, the needle carefully withdrawn and at the same time the puncture site covered with gauze. Occasionally, an unintentional premature pull of the needle may occur causing hematoma and pain.

Smaller needle size (measured in gauge, G) can be used to reduce the pain of the procedure [10], 11]. In a detailed trial of Arendt-Nielsen et al. with 30 healthy volunteers who were exposed to automated and standardized needle insertions (40–48 insertions per person) with 27G, 28G, 30G, 31G, and 32G needles, they found a significant and positive correlation between outer needle diameter and frequency of insertion pain [12].

For adults 21G needle size is commonly used, whereas 23G is used for aged patients, because of frequent poorer vein conditions. One standardly used regular 23G blood collection set is the Safety-Lok™ (SL) blood collection set with a standard thin-wall cannula and an inner diameter of 0.015 inches. Smaller needle sizes, like 25G needles, however, might impact blood quality as an important preanalytical requirement for laboratory measurements. Earlier studies reported that 25G needles are associated with increased hemolysis [13].

To facilitate easier blood collection especially regarding the above-mentioned needle draw challenges, a 25G blood sample set called UltraTouch™ Push Button (UT-PB) has been introduced. This set combines an ultra-thin wall needle, which reduces the cannula wall thickness and increases the inner diameter of the cannula to 0.016 inches (RightGauge™), with a specific cut needle tip (PentaPoint™) and automated needle retracing via button (Push Button). An increased rate of successful venipunctures with complex patient populations while statistically significantly reducing insertion force and overall pain during blood collection, without compromising efficiency and blood quality, was shown, potentially alleviating issues of small needle sizes associated with hemolysis [14].

Previous studies suggested that the specific needle tip geometry has impact on needed penetration force and therefore patients’ pain perception in non-blood taking aspects [15], [16], [17]. Comparing the specific 5-bevel design, which is used in the UT-PB set, and a common 3-bevel design, such as in the SL set, the 5-bevel design is supposed to significantly reduce needle insertion force compared to the common design [18].

While direct comparisons of different blood draw sets have previously been carried out for specific cohorts such as healthy volunteers [14], emergency room patients [19], cancer patients [20], and children [21], here we systematically focused on aged patients.

Materials and methods

Patients

A total of 161 inpatients with an age >60.0 years (range: 61.9–99.3 years with a mean of 81.6 years), treated in the Department of Geriatrics and Medical Gerontology at the Charité–Universitätsmedizin Berlin between July 2020 and May 2022, were enrolled prospectively. All patients reported informed consent. Blood sampling was performed in patients treated under routine clinical conditions. 89 patients underwent their regular blood collection performed with the SL collection set, 72 patients with the UT-PB set. Thirty-six patients underwent blood collection with both sets and on different days (Figure 1). In some cases, multiple rounds of blood collection were performed at different days leading to the final sample size as reported in Table 1.

Figure 1:

Flow chart with step-by-step process of the study. Graph depicts the step-by-step process of the study after inclusion.

Table 1:

Assessment of the phlebotomist.

Total cohort	Safety-Lok (n=120)	UltraTouch (n=120)	Difference, %	p-Value
Number of blood sample tubes	3.16 ± 0.5	3.29 ± 0.7	4.22	0.05
All planned tubes filled, %	99.17 ± 9.1	100.00 ± 0	0.84	0.32
All tubes filled completely, %	98.33 ± 12.9	97.50 ± 15.7	−0.85	0.65
Blood flow dry up, %	5.83 ± 23.5	5.00 ± 21.9	−14.29	0.78
Need to reposition the cannula to restore blood flow, %	8.33 ± 27.8	5.83 ± 23.5	−30.00	0.45
Immediate reddening (reaction of the skin), %	4.17 ± 20.1	6.67 ± 25.0	60.00	0.39
Hematoma at the injection site after 24 h or 48 h, %	30.83 ± 46.4	23.33 ± 42.5	−24.32	0.19
Size of hematoma, mma	19.30 ± 14.3	15.93 ± 13.9	−17.46	0.30
Severity of blood collection with the blood collection set in place with regard to finding veins and filling tubesc	1.38 ± 1.7	0.96 ± 1.4	−30.30	0.04
Sub-cohort	Safety-Lok (n=38)	UltraTouch (n=36)	Difference, %	p-Value
Number of blood sample tubes	3.16 ± 0.5	3.33 ± 0.7	5.56	0.13
All planned tubes filled, %	100.00 ± 0	100.00 ± 0	0	>0.99
All tubes filled completely, %	100.00 ± 0	100.00 ± 0	0	>0.99
Blood flow dry up, %	0.00 ± 0	0.00 ± 0	0	>0.99
Need to reposition the cannula to restore blood flow, %	0.00 ± 0	0.00 ± 0	0	>0.99
Immediate reddening (reaction of the skin), %	0.00 ± 0	5.56 ± 23.2	5.56	0.23
Hematoma at the injection site after 24 h or 48 h, %	26.32 ± 44.6	36.11 ± 48.7	37.22	0.45
Size of hematoma, mmb	17.70 ± 13.9	18.46 ± 16.3	4.30	>0.99
Severity of blood collection with the blood collection set in place with regard to finding veins and filling tubesc	1.26 ± 1.6	0.83 ± 1.4	−34.03	0.20

1. If not indicated differently, values represent mean ± SD, relative differences in percent, and p-value of Mann–Whitney test for statistical testing. Boolean data is presented as mean ± SD, in percent (%). ^an=37 (Safety-Lok) and n=28 (UltraTouch). ^bn=10 (Safety-Lok) and n=13 (UltraTouch). ^c0=none to 10=biggest conceivable.

Results

Assessment of phlebotomist

From the phlebotomist’s assessment of the total cohort as well as the cohort with both needle sets (sub-cohort) it was evident, that all intended blood tubes (three tubes were collected for both sets) could be successfully collected with both blood collection sets. Further, there were no significant differences in the completeness of their fill levels (Figure 2A and B, Table 1). In the total cohort (Figure 2A) applying the UT-PB blood collection set appeared to be 30 % easier (p=0.04) in locating veins and filling tubes. Nonetheless, both sets showed no significant difference regarding further procedural complications like dry up of blood flow or the need to reposition the cannula to restore blood flow during blood collection. However, the UT-PB set showed a trend toward less dry up (−14 %, p=0.78) and less repositioning (−30 %, p=0.45). Further, both sets showed no significant differences in the immediate skin reaction such as redness, appearance, or size of the hematoma after blood collection. Again, the UT-PB set tended to produce fewer (−24 %, p=0.19) and smaller hematoma (−17.5 %, p=0.30), but more immediate skin redness (60 %, p=0.39).

Figure 2:

Assessment of the phlebotomist. Graph depicts the difference between the two blood collection sets from total cohort (A) and sub-cohort (B) as relative change of UltraTouch set from Safety-Lok set in percent (red bars). Error bars represent standard deviation (SD). Numbers represent p-values of Mann–Whitney test.

In the sub-cohort, where both blood collection sets were applied (Figure 2B), there were no significant differences in the assessed parameters, with general improvement in blood collection (34 %, p=0.20), increase in skin reddening (6 %, p=0.23), and increase in hematoma appearance (37 %, p=0.45) as well as size (4 %, p>0.99) in those subjected to the UT-PB.

Subjective perception of patients

The assessment of the total cohort as well as the sub-cohort revealed no significant differences between the two sets used in regard to pain perception, fear, or the general assessment of the procedure (Figure 3A and B, Table 2). Nevertheless, there was a trend towards less pain perception using the UT-PB set (−16 %, p=0.30 (total cohort, Figure 3A); −30 %, p=0.39 (sub-cohort with both sets, Figure 3B)), whereas the blood collection was assessed as less pleasant with the UT-PB set (+16 %, p=0.77 (total cohort, Figure 3A); +25 %, p=0.8 (cohort with both, Figure 3B)).

Figure 3:

Subjective perception of pain. Graph depicts the difference between the two blood collection sets of total cohort (A) and sub-cohort (B) as relative change of UltraTouch set from Safety-Lok set in percent (red bars). Error bars represent standard deviation (SD). Numbers represent p-values of Mann–Whitney test. *Positive values represent an increase, negative values represent a decrease in pain and fear, respectively. **Positive values represent a decrease, negative values represent an increase in pleasantness, respectively.

Table 2:

Subjective patient’s perception.

Total cohort	Safety-Lok (n=120)	UltraTouch (n=120)	Difference, %	p-Value
Sensation of pain during blood samplinga	1.35 ± 1.7	1.13 ± 1.6	−16.05	0.30
Fear of the next blood samplea	0.65 ± 1.4	0.68 ± 1.7	2.53	0.35
General assessment of the blood collectionb	1.03 ± 1.4	1.20 ± 1.7	16.13	0.77
Sub-cohort	Safety-Lok (n=38)	UltraTouch (n=36)	Difference, %	p-Value
Sensation of pain during blood samplinga	1.26 ± 1.4	0.89 ± 1.1	−29.63	0.39
Fear of the next blood samplea	0.47 ± 1.0	0.31 ± 0.9	−35.49	0.33
General assessment of the blood collectionb	0.87 ± 1.1	1.08 ± 1.6	24.75	0.80

1. Shown values represent mean ± SD, relative differences in percent, and p-value of Mann–Whitney test for statistical testing. ^a0=none to 10=maximum; ^b0=pleasant to 10=unpleasant.

Discussion

In this study, we compared the routinely applied 23G 3-bevel needle with the smaller outer diameter (25G) 5-bevel UT-PB needle in an aged inpatient population addressing both clinical use and laboratory diagnostics. It is important to mention that the two sets differ in regard of needle outer diameter and the bevel of the needle, both characteristics could theoretically lead to lower pain perception in the UT-PB, with a smaller needle diameter and a flatter needle bevel.

We found no significant differences between both sets considering subjective perception of the patient such as pain, fear, and general perception of blood collection. Previous studies comparing different needle size in regard to pain found mixed results, ranging from no significant differences between different sized needles [10], 24], 25] to a significant decrease in pain perception with decreasing outer needle diameter [11], 12]. As stated above, there seems to be a correlation to some extent between needle tip bevel and pain. Hirsch et al. found a significant lower pain perception in 84 diabetes patients using the 5-bevel design compared to the 3-bevel design for repeated subcutaneous injections [16]. Interestingly, in the initial assessment of pain, where the patients were blinded for the used set, there was no difference documented in pain perception. The difference occurred only after repeated use and after informing the patients about the lower puncture force. Similarly, Jaber et al. reviewed five clinical trials comparing 5- to 3-bevel needles and found less pain occurrence with the 5-bevel set [26]. However, in our study we found no difference in pain perception between the 25G 5-bevel UT-PB and the 23G 3-bevel SL set. Our results are in line with a study conducted in 190 cancer patients, who showed no significant difference in pain perception between the UT-PB and the frequently used 23G needle [20]. In contrast, a study with 214 subjects (age ≥18 up to >70; 40 % healthy subjects, 60 % patients) reported significantly less specific pain during needle insertion with the UT-PB set compared to the SL set, but no significant difference in the overall pain assessment [14]. Accordingly, in a study with 420 children, particularly the age group 3–12 reported significantly less pain with the UT-PB set compared to SL [21]. One possible reason for these differing results could be that although pain perception varied between the two sets (−16 % in the total cohort and −30 % in the sub-cohort), the sample size of our study might have been underpowered and thus was too small to reach statistical significance. Another reason could be that sensitivity–in general-to low-intensity pain, as by blood draw, decreases with aging and our study population was the oldest of all studies cited [27], 28]. However, we did not observe a clear correlation of pain with age within our aged cohort (analyses not shown). Further, we found no difference between both sets used when considering fear of the next blood sampling, which is consistent with the results of existing studies [14].

In the total cohort, we observed less and smaller hematoma development in the study group punctured with UT-PB (−16 % and −22 %, respectively) compared to SL. However, the differences were not statistically significant and could not be reproduced in the sub-cohort when both sets were used (+37 % and +4 %, respectively). In comparison to our results, Imbrahim et al. found in 50 volunteers, who underwent arterial puncture with either an 29G or a 23G needle, significantly less complications like swelling or bruising in the smaller needle group [29]. Another study reported no difference in hematoma size and pain of injection between 30G and 26G subcutaneous needles in 64 patients [24]. It is noticeable that studies with larger size needles rather show differences in hematoma development in contrast to studies with smaller size needles, which show no differences. It is conceivable that the effect is linked to larger sizes and diminishes with small sized needles, which, however, questions the major impact on a patient individual level.

In regard to laboratory measurements particularly focusing on hemolysis susceptibility, we did not find clinically relevant differences in any parameter between blood sample sets used, a result which is also consistent with other studies [13], 14], 30]. Importantly, direct comparisons, e.g., concerning non-hemolysis-susceptible analytes, should be regarded with caution since blood collection and laboratory analysis were not performed on the same day. Thus, changes within the time of hospitalization could impact measured values. The hemolysis index, indicating the grade of erythrocytes damage and thus the level of free blood hemoglobin, was reduced about 28 % in the total cohort and 47 % in the sub-cohort, although statistically not significant. Here, repeating the study with a larger cohort would possibly unravel, if indeed the UT-PB or comparable sets would lead to less hemolysis. However, ranges of hemolysis indices detected here, are still below relevant interference with laboratory analysis methods and without clinical impact for hemolysis-sensitive parameters.

From the phlebotomist’s subjective assessment, we found a significant difference in favor of the UT-PB set compared to the SL set regarding general severity of blood collection (30 %). This is comparable to the results of Hirsch et al., where diabetes patients favored the 5-bevel set in respect to comfort and ease of insertion [16] and to the results of Padoan et al., where phlebotomists rated the UT-PB set used in children associated with lower difficulty in general compared to the SL set [21]. However, it must be emphasized that this represents a subjective parameter that cannot be blinded due to the technical implementation and could therefore be biased.

Based on previous observations in a pediatric cohort (n=230 children) analyzing pain perception at the age group of 3–12 years [21], we included a comparable cohort size in our study. Although in general not underpowered, a major limitation of the study is the small size of the sub-cohort in which the same patients had blood drawn with both devices. The current sub-cohort, however, did not provide relevant differences in parameters, e.g., pain perception, between both systems. Furthermore, as the UT-PB-system combined a flatter needle bevel and a small needle diameter, the study cannot distinguish the influence of these features on pain perception. Another limitation is that blood draw was performed only by a single phlebotomist. Further, information on current pain medication treatment was not collected, leading to uncertainty about difference in such treatment between groups. However, we specifically aimed to conduct our study in a realistic non-homogenous cohort and therefore, other than age, did not introduce any exclusion criteria, like medication and comorbidities, in our study.

Taken together, there is little evidence from the results of our study that a 25G 5-bevel blood collecting set such as the UT-PB is associated with improved, significantly better handling compared to a commonly and routinely used set in an aged population. New medical devices such as the UT-PB could presumably help in treatment and reduce complications particularly at an individual level in aged patients, when more difficult blood access is evident.