Development of Guidelines for Accurate Measurement of Small Volume Parenteral Products Using Syringes

Abstract

Background: Syringes are commonly used in pharmacy compounding for the measurement of small volumes, especially in the preparation of sterile products for injection and infusion. However, there are no current official guidelines for the proper use of syringes in measuring small volumes. Objective: The purpose of this project was to determine the accuracy and precision of commercially available syringes in measuring small volumes during sterile product preparation to make recommendations for syringe size selection. Methods: To assess precision and accuracy of syringes, 3 separate investigators measured 5%, 10%, or 20% (n = 30 each) of the volume of a 1-, 3-, 5-, 10-, or 20-mL syringe with an attached 18G, 1½” needle by drawing sterile water for injection from a vial. Delivered volumes were measured gravimetrically using an electronic balance and converted to volume using the specific gravity of water (1.0). Accuracy is represented as the mean and standard deviation, while precision is represented as percent relative standard deviation. Differences were assessed using a 1-way analysis of variance with Bonferroni adjustments and significance set at P < .05. Results: Precision and accuracy were highly variable and often significantly (P < .05) different compared to the theoretical volume delivered both within and between investigators. An increased likelihood of unacceptable error (>5%) was observed when less than 20% of the labeled capacity of a syringe was measured. Mean percent error ranged from 1.4% to 18.6%, despite manufacturer specification of ±5% accuracy, suggesting proper technique as a major factor in small-volume measurements. Conclusion: In addition to proper, validated training of syringe users, we recommend that users measure no less than 20% of the indicated volume of the syringe while choosing syringes as close as possible to the desired measurement. When possible, very small volumes should be diluted to meet the minimum volume of the smallest syringe available. Implementation of these recommendations will improve accurate dosing and, ultimately, patient safety.

Introduction

Syringes are commonly used in pharmacy compounding for the measurement of small volumes, especially in the preparation of sterile products for injection and infusion. The measurement of some medications—for example, potent drugs, drugs with narrow therapeutic margins, ophthalmic preparations, and pediatric dosing—requires a high degree of precision and accuracy for patient safety.

Currently, no official guidelines exist for either minimum or maximum volumes for accurate measurement using syringes. The general rule of thumb for measuring bulk liquids using volumetric glassware is to measure at least 20% of the total volume of the measuring device while also using the smallest instrument possible to make the measurement.¹ Clinically, unsubstantiated recommendations for measurement of small volumes in syringes varies including ranges from 20%, similar to the guidelines for volumetric glassware, up to a minimum volume of 50% of the syringe capacity. These values differ significantly from the manufacturer specifications, which state a volumetric accuracy for some syringes of ±5%, in line with current ISO standards which require a tolerance no greater than ±5% at more than or equal to 50% capacity (Tables 1 and 2).^2,3 In addition to the inherent variability in instrumentation, multiple other factors can contribute to inaccurate and imprecise measurements including user skills and the volume to be measured relative to syringe capacity.

Table 1.

Manufacturer’s Specifications for Syringes and Needles.²

BD syringe size		1 mL	3 mL	5 mL	10 mL	20 mL
	Major graduation (mL)	0.1	0.5	1.0	1.0	5.0
	Minor graduation (mL)	0.01	0.1	0.2	0.2	1.0
	Dead space (mL)	0.07	0.07	0.075	0.10	0.15
	Volumetric accuracy	±5%a	±5%	±5%	±5%	±5%

Note. BD = Becton Dickinson.

^aFor 1 mL syringe, the accuracy below 0.2 mL is 0.07 mL.

Table 2.

Measured Volumes Based on Nominal Syringe Capacity in Milliliters (mL), Including Expected Syringe Tolerance Per ISO 7886-1:2017.^2,3

Nominal volume (Vnominal)	5% capacity			10% capacity			20% capacity
	Expelled volume (mL) (Vexpelled)	Tolerance (mL)a	Tolerance (%)	Expelled volume (mL) (Vexpelled)	Tolerance (mL)a	Tolerance (%)	Expelled volume (mL) (Vexpelled)	Tolerance (mL)a	Tolerance (%)
1 mL	0.05	0.016	32	0.1	0.017	17	0.2	0.019	9.5
3 mL	0.15b	0.048	32	0.3	0.051	17	0.6	0.057	9.5
5 mL	0.25b	0.078	31.2	0.5b	0.08	16	1	0.085	8.5
10 mL	0.5b	0.155	31	1	0.16	16	2	0.17	8.5
20 mL	1	0.31	31	2	0.32	16	4	0.34	8.5

^aTolerance (mL) = (1.5% V_nominal) + 2% V_expelled for V_nominal less than 5 mL; tolerance (mL) = (1.5% V_nominal) + 1% V_expelled for V_nominal more than 5 mL.

^bDue to increments of the indicators on these syringe, the drawn volumes were estimated between 2 graduation markings.

Few studies have attempted to address the issue of using syringes to measure small volumes during compounding. Several studies that have been published specifically refer to the accuracy of insulin measurements in a 1-mL syringe.^4-6 Two published studies have looked at the accuracy of measurements in a variety of syringes but the results did not reach statistical significance to make a strong recommendation on compounding practices.^7,8 Lee et al⁷ concluded that reproducibility increased as the measured volume increased relative to syringe capacity and that, relative to the measured volume, smaller syringes yielded more reproducible results. Similarly, Erstad et al⁸ found that measurement of 0.5 mL in 3 different syringe sizes (1, 3, and 10 mL) was more accurate and reproducible in the smaller syringe.

Therefore, it seems necessary to develop guidelines for the proper use of syringes, especially in measuring small volumes and potent medications, to improve dosing accuracy and ultimately patient safety. Scientific evidence of the limits of accuracy should be used by pharmacists and other health care providers in medication preparation and dosing to ensure optimal patient outcomes.

Methods

Materials

Syringes (1, 3, 5, 10, and 20 mL) and 18G, 1½” needles were purchased from Beckton-Dickinson (Franklin Lakes, NJ). Sterile water for injection (SWFI), 10-mL vials were purchased from Hospira (Lake Forest, IL).

Volumetric Measurement

An 18G, 1½” needle was attached to a 1-, 3-, 5-, 10-, or 20-mL syringe, and the syringe assemblage, with needle cap removed, was weighed (W_syringe) using a calibrated analytical balance accurate to ±0.001 g (Denver Instruments, Bohemia, NY). A volume of SWFI equivalent to 5%, 10%, or 20% of the indicated nominal capacity of the syringe (Table 2) was drawn into the syringe, ensuring all air bubbles were removed, and the syringe assemblage was weighed again (W_filled). The contents were then expelled from the syringe and the syringe assemblage was reweighed (W_expelled). Three separate investigators repeated each measurement 30 times.

Data Analysis

The weight of water delivered from the syringe was calculated by subtracting the weight of the syringe assemblage after the volume was expelled from the weight of the syringe assemblage containing the desired volume of water (5%, 10%, or 20% NC; Table 2):

$${\mathrm{W}}_{\mathrm{delivered}}{=\mathrm{W}}_{\mathrm{filled}}-{\mathrm{W}}_{\mathrm{expelled}}$$

All weights were converted to volume, in milliliters (mm), using the density of SWFI at room temperature, which was assumed to be 1.0 g/mL.

The accuracy of the measurements was calculated as the mean percentage error (MPE) using the mean and standard deviation of the percentage error (%E) between the expected volumes (Table 2) and the delivered volumes:

$$\begin{array}{l}\%\mathrm{E}=\frac{{\mathrm{W}}_{\mathrm{delivered}}-{\mathrm{W}}_{\mathrm{expected}}}{{\mathrm{W}}_{\mathrm{expected}}}\times 100\\ \mathrm{MPE}=\frac{\sum \%\mathrm{E}}{\mathrm{n}}\left(\mathrm{where}\mathrm{n}=\mathrm{number}\mathrm{of}\mathrm{measurements}\right)\end{array}$$

Reproducibility was calculated as the percent relative standard deviation (%RSD):

$$\%\mathrm{RSD}=\frac{\mathrm{SD}}{\overline{X}}\times 100$$

A 1-way analysis of variance with Bonferroni adjustments and significance set at P < .05 was used to assess variability between syringe measurements and between investigators.

Results

Accuracy

For all syringe sizes tested, the accuracy improved as the percentage of the nominal volume measured increased (P < .005; Figure 1). Additionally, the total number of measurements with greater than 5% error decreased while the percentage of measurements with an acceptable error (≤5%) increased with increasing percentage nominal capacity measured (Figure 3). Comparison between different syringe sizes showed wide variability in accuracy. Generally, the accuracy of measurements improved as syringe size increased except for the 5-mL syringe. The MPE varied significantly between syringe sizes, the percentage (5%, 10%, or 20%) of the instrument capacity measured, and the investigator performing the measurements (Figure 1).

Figure 1.

Reproducibility and accuracy of syringe measurements.

Note. (a) Reproducibility of measurements, represented as percent relative standard deviation (%RSD) and (b) accuracy of measurements, represented as mean percent error ($\overline{X}$ ± SD) using syringes drawn to 5%, 10%, or 20% of capacity. Results represent pooled data taken by 3 separate investigators (n = 90).

^aMeasurements estimated between syringe graduation marks.

*P < .05 for 20% capacity measures relative to 5% measures. **P < .05 for 20% capacity measures relative to both 5% and 10% within syringe measures.

Figure 3.

Measurement error.

Note. Percentage of measurements (pooled data; n = 90) with acceptable error (≤5%, ■), more than a 5% error (⍰), and more than 10% error (). Category X-axis labels indicate syringe size and percentage of syringe capacity measured.

Significant differences (P < .05) were observed in the accuracy between the 3 investigators especially with smaller volumes, with 2 of the investigators tending to over-measure while the third was usually within 5% error range (Figure 2). However, as the syringe size increased, measurements tended to approach the desired volume for all investigators, especially as percentage of the nominal capacity measured increased. Additionally, the difference between investigators also decreased as syringe size increased, especially at the 5% capacity measurement.

Figure 2.

Box and whisker plots representing the volume measured in each syringe size at 5%, 10%, and 20% capacity, respectively, for each of 3 investigators (A ⍰ B C ■).

Note. Box represents first quartile, median, and third quartile. Whiskers represent 1.5-fold of the quartiles while individual points represent outliers. Dashed lines indicated intended measured volume.

^aData for 5% capacity, 3-mL syringe is missing for Investigator A.

*P < .05.

Reproducibility

For each of the syringe sizes tested, the %RSD decreased as the percentage of the measured syringe capacity increased toward 20%, although the results were generally not statistically significant (P > .05) except for the 5-mL syringe (Figure 1a). This trend was observed for the overall pooled data as well as for each individual investigator. The average %RSD, reflecting the reproducibility of the measurements, was below 5% for each syringe tested at 20% capacity (Figure 1a). However, the accuracy of the measurements was frequently greater than 5% and varied significantly between investigators, suggesting the technique as an important variable in measurement accuracy (Figures 1b and 2).

Discussion

Technical data provided by the manufacturer indicate tolerance within ±5% for the syringes used in this study (Table 1).² Sánchez-Rubio Ferrández et al⁹ investigated the error associated with the measurements at full syringe capacity in 10-, 20-, and 50-mL syringes and found that in all cases, the error was less than 5%, which is in line with the inherent variability of accuracy as stated by the manufacturer. Similarly, Lee et al⁷ measured the accuracy and reproducibility of 2 brands of 1-, 3-, 10-, and 12-mL syringes (primarily >20% nominal capacity) and found that the MPE (accuracy) was less than 5% and the %RSD (reproducibility) was less than 1% for all measurements. Although we investigated only one brand, it is expected that syringes supplied by other manufacturers should have the same tolerance because all syringes are manufactured to the same ISO standards (ISO 7886-1-2017). Per ISO 7886-1³ standards, the tolerance of all syringes when measuring less than 50% of nominal capacity (V_nominal) is ±[1.5% V_nominal + 2% V_expelled] for syringes with a nominal capacity less than 5 mL and ±[1.5% V_nominal + 1% V_expelled] for syringes more than 5 mL. Tolerances of ±5% or less are only required when the expelled volume is at least 50% of the nominal volume. However, manufacturers may employ more stringent tolerances to their product, as with the Becton Dickinson (BD) syringes used in the study, for which the manufacturer states a ±5% tolerance without regard to volume except for the 1-mL syringe.² It is important to note that the tolerances for syringes refer only to the syringe and not to its use with an attached needle or injection via a Leur-Lok connector on an intravenous administration set.

At partial capacity (5%-20%) volumes used in this study, a range of accuracy of 1.4% to 18.6% was observed, suggesting that factors beyond the acceptable syringe tolerance contributed to mismeasurement. In this study, both accuracy and reproducibility improved as both syringe size and the percentage of nominal capacity measured increased. However, measurement error was frequently greater than 5% (Figure 3), suggesting that the technique also played a significant role in measurement accuracy. Mean percentage error tended to decrease as syringe size increased except for the 5-mL syringe, possibly due to the graduations on the syringe barrel, which fell between graduation marks for the 5% and 10% capacity measures. Thus, these volumes were estimated measurements, contributing to decreased accuracy (up to 20%; Table 2). However, the error in this syringe at 20% nominal capacity, which fell directly on a graduation mark, was still high relative to other syringes tested. Additionally, the 5% capacity measures for both the 3- and 10-mL syringes also fell between the graduation marks but did not appear to have as significant an impact on the overall accuracy as compared to the 5-ml syringe. Thus, estimation between graduation marks may not be as much of a contributor to overall measurement error as appropriate instrument selection and compounder technique. Indeed, a significant difference in error was observed between the 3 investigators in this study for most of the syringe measurements (Figure 2).

Overall, reproducibility improved as syringe size decreased relative to the measured volume, which is consistent with the literature. Thobani and Steward¹⁰ measured volumes of 1, 3, or 5 mL in syringes ranging from 1- to 20-mL capacity (percentage of nominal capacity measured ranged from 5% to 100%). The authors found that although less than 5% in all cases, the MPE tended to be higher when less than 20% of the nominal capacity was measured (1 mL measured in a 10- or 20-mL syringe). Reproducibility also decreased as the syringe size increased, which the authors attributed to increased internal diameter of the barrel.

Erstad et al hypothesized that using large-volume syringes to measure small volumes would result in increased error. The authors measured 0.5 mL in 1-, 3-, or 5-mL syringes (10%-50% nominal capacity) and found that as syringe size increased, reproducibility and accuracy decreased.⁸ Accuracy was acceptable (<5%) for only the 1- and 3-mL syringes, while reproducibility fared worse with an acceptable error observed only for the 5-mL syringe. Similar to Thobani and Steward,¹⁰ the authors hypothesized that the increase in internal diameter for larger syringes made accurate measurement more difficult in addition to human perception of the plunger position. In another study, Raju and Weinberg¹¹ measured 0.05 mL (5% capacity) and 0.10 mL (10% capacity) doses in 1-mL syringes. The authors found that for the smaller volume, 22% of the measurements had more than 20% error and only 44% of the measurements were within 10% of the desired volume. As the volume increased to 10% of syringe capacity, 71% of the measurements were within 10%, with only 7% having more than a 20% error.

Casella et al⁴ investigated the accuracy and precision of low-dose insulin syringes by measuring 0.5 units (0.005 mL) to 2.0 units (0.02 mL) in 0.3- or 0.5-mL insulin syringes. The authors found that an unacceptable error (>10%) was achieved when less than 2.0 units of insulin were measured regardless of syringe size and that doses tended to be over-measured in all cases. Because of the inaccuracy of low-dose insulin measurements and the lack of availability of more precise measuring devices, the authors recommended dilution of the drug product for all doses less than 2.0 units (6.7% and 4% of nominal capacity, respectively, for 0.3 and 0.5 mL syringes). Keith et al¹² measured 1 (0.01 mL), 2 (0.02 mL), and 5 (0.05 mL) unit insulin doses in 2 brands of 30-unit insulin syringes, equivalent to 3.3% to 16.7% nominal capacity. Doses were generally over measured, and accuracy was poor (mean error >10%) for volumes less than 10% nominal capacity, and similar to Casella et al, the authors recommend predilution of insulin before low-dose measurement. However, both precision and accuracy improved as the volume measured increased, with less than 5% error for the 5-unit dose. Similarly, Lteif and Schwenk¹³ found that when patients were asked to measure low-dose insulin (≤5 units) using 0.3- or 0.5-mL insulin syringes, the mean error in the measurements was 9.9% ± 2.4%.

Other contributing factors to mismeasurement that have been identified in the literature include size of dead space relative to the syringe capacity, calculation errors, infrequent use of syringes in daily routine, number of years of professional experience, and fatigue.^14,15 Each of these factors can be linked back to appropriate didactic and practical training skills necessary for preparing competent sterile products.

Conclusion

Proper measurement of liquid dosage forms using syringes is imperative for accurate dosing of medication. Evidence from this study as well as a review of the current literature can be used to compile a list of recommendations for the effective use of hypodermic syringes as measuring devices:

• Select syringes as close as possible in volume to the desired measurement.²

• Select the smallest syringe possible such that more than 20% of the nominal syringe capacity is measured to limit instrumental error.^7,8,11

• Prepare a dilution of the solution being measured, if possible, when the desired volume is less than 20% of the smallest syringe available.^4,7,16

• Properly train, certify, and periodically recertify (eg, annually) personnel in didactic and practical skills necessary to proper use of syringes.¹⁴

Implementation of these recommendations will minimize errors in dosing and ultimately improve patient safety.