Infusion pressure and pain during microneedle injection into skin of human subjects

Abstract

Infusion into skin using hollow microneedles offers an attractive alternative to hypodermic needle injections. However, the fluid mechanics and pain associated with injection into skin using a microneedle have not been studied in detail before. Here, we report on the effect of microneedle insertion depth into skin, partial needle retraction, fluid infusion flow rate and the co-administration of hyaluronidase on infusion pressure during microneedle-based saline infusion, as well as on associated pain in human subjects. Infusion of up to a few hundred microliters of fluid required pressures of a few hundred mmHg, caused little to no pain, and showed weak dependence on infusion parameters. Infusion of larger volumes up to 1 mL required pressures up to a few thousand mmHg, but still usually caused little pain. In general, injection of larger volumes of fluid required larger pressures and application of larger pressures caused more pain, although other experimental parameters also played a significant role. Among the intradermal microneedle groups, microneedle length had little effect; microneedle retraction lowered infusion pressure but increased pain; lower flow rate reduced infusion pressure and kept pain low; and use of hyaluronidase also lowered infusion pressure and kept pain low. We conclude that microneedles offer a simple method to infuse fluid into the skin that can be carried out with little to no pain.

Figure 1

Needles used for intradermal infusion. Microneedles measuring (a) 500 μm, (b) 750 μm, (c) 1 mm and (d) 4 mm in length. (e) Hypodermic needle with intradermal bevel (26 gauge). All images are shown at the same magnification and obtained by brightfield microscopy.

Figure 2

Effect of microneedle insertion depth on (a) infusion pressure and (b) pain score as a function of infusion volume. Microneedles were inserted into the skin at T1 = 500 μm (▲), T2 = 750 μm (■) and T3 = 1 mm (○). A microneedle was inserted into the subcutaneous space at T4 = 4 mm (△). A hypodermic needle T5 (●) was also inserted intradermally. All infusions were performed at the medium saline flow rate of 0.3 mL/min. Visual analog score (VAS) for pain was normalized relative to the pain caused by intradermal insertion of the hypodermic needle. Data are the average of 10 replicates expressed as mean ± SD.

Figure 3

Effect of microneedle retraction on (a) infusion pressure and (b) pain score as a function of infusion volume. Microneedles were inserted into the skin at T1 = 500 μm (▲), T3 = 1 mm (○) and T6 = 1 mm followed by retraction back to 500 μm (◇). A hypodermic needle T5 (●) was also inserted intradermally. All infusions were performed at the medium saline flow rate of 0.3 mL/min. Visual analog score (VAS) for pain was normalized relative to the pain caused by intradermal insertion of the hypodermic needle. Data are the average of 10 replicates expressed as mean ± SD.

Figure 4

Effect of infusion flow rate on (a) infusion pressure and (b) pain score as a function of infusion volume. Microneedles were inserted into the skin at 750 μm and saline was infused at rates of T8 = 0.1 mL/min (◆), T2 = 0.3 mL/min (■), and T7 = 1 mL/min (□). Visual analog score (VAS) for pain was normalized relative to the pain caused by intradermal insertion of the hypodermic needle. Data are the average of 10 replicates expressed as mean ± SD.

Figure 5

Effect of hyaluronidase on (a) infusion pressure and (b) pain score as a function of infusion volume. Microneedles were inserted into the skin at 750 μm and T2 = saline (■) and T9 = hyaluronidase (*) were infused at a rate of 0.3 mL/min. Visual analog score (VAS) for pain was normalized relative to the pain caused by intradermal insertion of the hypodermic needle. Data are the average of 10 replicates expressed as mean ± SD.

Figure 6

Absolute VAS pain scores associated with insertion of microneedles into skin. Pain scores are reported on a 100-mm VAS scale for insertion of microneedles to 500 μm, 750 μm, 1 mm and 4 mm depths and to 1 mm depth followed by 500 μm retraction. A 26 gauge hypodermic needle was also inserted. Data are the average of 10 replicates expressed as mean ± SD, except for 750 μm, where pain scores from T2, T7, T8 and T9 were averaged based on 40 replicates. * p < 0.05.

Figure 7

Scatter plot representation of the normalized VAS pain scores for all experimental conditions as a function of infusion pressure. See Table 1 for definitions of groups. Data are obtained from Figs. 2 – 6 and represent the average of 10 replicates expressed as mean ± SD.

Figure 8

Overall perception of pain for the nine treatments (as defined in Table 1). Subjects were asked to rate the overall pain associated with insertion and infusion for each treatment as: no pain, no pain to mild pain, mild pain, mild to moderate pain, moderate pain, moderate to severe pain, severe pain, or worst possible pain. Each bar represents the input from 10 subjects.

Figure 9

Photographs and ultrasound echographs of skin before and after microneedle-based infusion. For each treatment (as defined in Table 1), the upper pair of images displays a photograph of the skin surface (on the left) and an echograph of the skin thickness (on the right) taken before infusion. The lower pair of images displays a photograph of the skin surface immediately after infusion (on the left) and an echograph of the skin thickness ~5 min after infusion (on the right). The bright white line on the top of the echographs represents the epidermis; the green layer below it depicts the dermis; and the black region below the dermis is hypodermis. The regions of red and yellow in the dermis represent areas of higher acoustic reflection corresponding to denser tissue. Red arrows indicate black space between epidermis and dermis with low acoustic reflection, which are believed to correspond to saline pools. Yellow arrows indicate black spaces corresponding to possible saline pools within the dermis.