Article, Endocrinology

Rapid vascular glucose uptake via enzyme-assisted subcutaneous infusion: Enzyme-Assisted Subcutaneous Infusion Access Study

Original Contribution

Rapid vascular glucose uptake via enzyme-assisted subcutaneous infusion: Enzyme-Assisted

Subcutaneous Infusion Access Study?,??,?

Olanrewaju A. Soremekun MD a,b, Melissa L. Shear BA b, Sagar Patel BS b, Gina J. Kim BS b, Paul D. Biddinger MD b, Blair A. Parry BA, CCRC b,

Maria A. Yialamas MD c, Stephen H. Thomas MD, MPH b,?

aHarvard Affiliated EM Residency Program, Brigham and Women’s Hospital and Massachusetts General Hospital,

Boston, MA, USA

bDepartment of Emergency Services, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA

cDepartment of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA

Received 1 August 2008; revised 26 August 2008; accepted 28 August 2008


Background: Enzyme-assisted subcutaneous infusion (EASI), with subcutaneous human recombinant hyaluronidase pretreatment, may offer an alternative to standard Intravenous access.

Objectives: This study’s objectives were to assess paramedic (Emergency Medical Technician- Paramedic [EMTP])-placed EASI access in volunteers to determine (1) feasibility of EMTP EASI access placement; (2) subject/EMTP ratings of placement ease, discomfort, and overall EASI vs IV preference; and (3) speed of intravascular uptake of EASI infusate.

Methods: Twenty adults underwent 20-gauge IV placement by 4 EMTPs, receiving a 250-mL maximal- rate IV bolus of normal saline. Next, each subject received in the other arm a 20-gauge EASI access line (with 1-mL injection of 150 U of human recombinant hyaluronidase), through which was infused 250 mL D5NS (1 g glucose was labeled with stable tracer 13C). Blood draws enabled gas chromatography/mass spectrometry (GC/MS) assessment of 13C-glucose uptake. Intravenous access and EASI access were compared for time parameters and subject/EMTP ratings. Data were analyzed with median and interquartile range, Kruskal-Wallis testing, Fisher exact test, and regression (GC/MS data).

Results: Intravenous access and EASI access were successful in all 20 subjects. Compared with EASI access (all placed in b15 seconds), IV access took longer; but the 250-mL bolus was given more quickly via IV access. EMTPs rated EASI easier to place than IV; pain ratings were similar for IV and EASI. The GC/MS showed intravascular uptake at all time points.

? Grant support: Project supported by an unrestricted grant from Baxter, the company that markets the human recombinant hyaluronidase used in the study.

?? Trial registration:, NCT00386386.

? Previous presentation: Abstract presentation at the 2008 Society for Academic Emergency Medicine Annual Meeting, 29 May-1 June, Washington DC.

* Corresponding author. MGH Department of Emergency Services; Zero Emerson Place Suite 3B; Boston, MA 02114-2241, USA. Tel.: +1 617 726 7622;

fax: +1 617 724 0917.

E-mail address: [email protected] (S.H. Thomas).

0735-6757/$ – see front matter (C) 2009 doi:10.1016/j.ajem.2008.08.028

Conclusions: Enzyme-assisted subcutaneous infusion is faster and easier to initiate than IV access; intravascular absorption of EASI-administered fluids begins within minutes.

(C) 2009


Administration of fluids in the prehospital setting remains an important intervention in prehospital and disaster circumstances [1-3]. Even when there are sufficient Advanced life support -level responders, the out- of-hospital setting can pose challenges to intravenous catheter placement. These challenges range from anatomy (eg, venous collapse in hypotension) to positioning (eg, entrapment) and environmental conditions (eg, poor lighting, vehicular motion). Problems are compounded in multipatient situations such as Mass casualty incidents .

Even if IV access placement were always achievable, there are reasons for interest in alternative mechanisms for delivery of parenteral fluids. Prolonged placement times and need for multiple IV “sticks” before successful IV cannula- tion cause considerable discomfort in neonates, children, and adults [4-9].

Currently, there is only 1 non-IV choice–the intraoss- eous (IO) line–for prehospital access to the intravascular compartment. Intraosseous lines have an acknowledged role in Prehospital medicine, but can be time consuming and painful (and even cause fracture); and significant initial and ongoing skills maintenance is required [10].

Thus, there is potential utility for an alternative means for gaining access to the intravascular compartment, if such a new mechanism is easily taught, quickly performed, relatively painless, and reliable in its delivery of infusate to the vascular space. For relevance to prehospital and MCI scenarios, ALS providers should be able to use the alternative access approach; ideally, such an approach should be feasible for use by Basic Life Support responders.

This study focuses on the alternative parenteral fluid delivery mechanism of subcutaneous (SC) infusion, facili- tated by preadministration of human recombinant hyalur- onidase (HRH). The HRH hydrolyzes hyaluronan, which constitutes the major SC diffusion barrier in the gel-like extraadipocyte portion of the interstitial matrix. Although hyaluronan is found in lower concentration than collagen in the skin, it plays a disproportionately large role in resisting fluid movement [11].

Pretreatment with HRH (Food and Drug Administra- tion-approved in 2005) as part of an approach called en- zyme-assisted SC infusion (EASI) access increases local dispersion and absorption of SC-administered drugs and fluids. This is achieved by hyaluronidase’s modification of connective tissue permeability via hydrolyzing hyaluronic acid, effecting a cleavage of the glucosaminidic bond between N-acetylglucosamine and glucuronic acid moieties. The cleavage results in a decrease in viscosity of the

cellular cement and promotes diffusion of injected fluids, facilitating their absorption. The decrease in viscosity is reversed within 24 hours because of the rapid inactivation of the hyaluronidase enzyme and also because of the rapid turnover rate of skin hyaluronan [12-14].

So-called spreading agents, historically derived from animal extracts, have been used clinically to facilitate dispersion and absorption of other drugs for more than 50 years [12]. The Food and Drug Administration approval for HRH states the drug is “indicated as an adjuvant to increase the absorption and dispersion of other injected drugs, for hypodermoclysis, and as an adjunct in subcutaneous urography for improving resorption of radiopaque agents” [15]. A previous volunteer study has demonstrated that, with use of a 24-gauge catheter, isotonic solution (lactated Ringer solution) can be infused with no discomfort at maximal gravity-assisted rates exceeding 500 mL/h [16]. The findings of complication-free infusion rates of a variety of fluids (normal saline [NS], D5NS, D51/2NS, D5W) in the range of 40 to 500 mL/h have also been confirmed in patients in the hospice setting [17].

The concept of EASI access has been demonstrated to be feasible, with preliminary work concentrating in the hospital and hospice settings [12,18]. The current investigation aimed to assess EASI access in a simulated prehospital/MCI setting. This study’s primary goals were to assess the feasibility of ALS provider placement of EASI access lines and to determine whether EASI-administered infusate actually reached the intravascular compartment. The latter goal was addressed by assessing serial (q15-minute) blood samples from the arm contralateral to EASI access infusion of tracer- labeled glucose. The overall feasibility assessment included investigation of whether EASI access and IV access were

similar with respect to operator and subject preference.



The study was conducted in the Emergency Department of the Massachusetts General Hospital (MGH), where Institutional Review Board approval was obtained.


The 4 ALS providers responsible for placing EASI and IV access were paramedics (Emergency Medical Technician- Paramedic [EMTP]), all actively involved in prehospital

care. All 4 had at least 3 years’ experience at the EMTP level (range, 3-10 years). None of the study ALS providers had any familiarity with, or opinions regarding, EASI access or SC fluids administration before the study. The morning of the study, EMTPs underwent 5 minutes’ EASI access training by a registered nurse employed by the company marketing the HRH and sponsoring the study (Baxter Healthcare, Chicago, IL). For safety reasons, this registered nurse (who had extensive experience in EASI access) was present during the study; but she did not place any EASI access lines and did not intervene in the care of any study subjects.

The 20 study participants were Healthy adults who responded to an announcement requesting participants in a disaster drill-type exercise with the main focus being assessment of EASI access. A total of 25 respondents underwent informed consent, with an a priori study accrual plan for actual intervention in 20 subjects. Study participants had not heard of, nor had they any opinions regarding, EASI access. Most study participants were members of the MGH disaster response team; others included hospital-affiliated nonclinical staff.

Subjects, who were at least 18 years of age (median age, 43; range, 21-54; interquartile range [IQR], 30-51) and 55% male, had none of the following conditions: pregnancy, diabetes, or upper extremity acute medical or surgical disease. Subjects could have no history of Severe asthma, venous thrombosis, or renal or cardiac disease. Subjects could not be anticoagulated or be taking steroids or other immunosuppressants. Because of the potential for reduced hyaluronidase effectiveness [15], the study excluded patients taking more than 80 mg daily aspirin, as well as any patients taking adrenocorticotropic hormone, antihistamines, or estrogen other than in oral contraceptive preparations. Subjects were not required to fast before the study, but were not allowed to eat or drink during the EASI infusion or the phlebotomy sampling time frame.


Study subjects underwent a single upper extremity 20- gauge IVaccess placement in the site identified as best by the EMTP operator (either arm could be used). Subjects were requested to indicate, on a 0-to-10 Numeric rating scale (NRS), the pain associated with IV placement. The EMTPs were asked to indicate, on a 5-point Likert scale, the difficulty of IV placement. A bolus of 250 mL NS was infused at maximal rate, and subjects were prompted to indicate pain levels at the conclusion of the IV infusion.

After completion of the NS infusion, subjects underwent placement of a 20-gauge EASI access catheter (Fig. 1). The EASI access catheter was placed in the arm contralateral to that in which IV access was placed. In 12 subjects (2 of whom were left-hand dominant), EASI access was estab- lished in the left arm; in 8 subjects (1 of whom was left-hand dominant), EASI access was established in the right arm. As part of establishment of EASI access, 150 U of HRH

(Hylenex, Baxter) was injected in a 1-mL volume. Upon placement of the EASI access line, subjects rated discomfort (on the NRS) and EMTPs rated placement ease (Likert scale). Subjects’ extremity circumference in the arm receiving the SC infusion was assessed before EASI access placement and upon completion of the EASI infusion.

To track movement from the SC compartment to the intravascular space, the EASI-administered infusates were different from the NS given in the IV bolus. A total of 12.5 g glucose, 1 g of which was stable tracer-labeled with 13C (see next section for details), was added to NS to constitute D5NS. A bolus of 250 mL of the tracer-labeled D5NS was administered at a rate initially determined by patient comfort and then set at approximately 100 mL/h (see “Results” section for explanation of rate adjustment).

Just before starting the D5NS infusion in the EASI access line, a baseline (time 0 minute or t0) blood sample was taken from the indwelling (heparin-locked) IV line. Samples of blood were drawn from a subject’s IV line at times t = 15, 30, 45, and 60 minutes (t15, t30, t45, and t60, respectively) post-EASI infusion commencement. Blood was only drawn from the indwelling IV line, and only 3 attempts were made (per subject) at each time frame; if the 3 attempts at phlebotomy failed, that timed phlebotomy was aborted. The blood samples were stored on ice for 3 to 5 hours before transport to a university laboratory (Massachusetts Institute of Technology [MIT]) next to the MGH. Methodological detail pertinent to gas chromatography/mass spectrometry (GC/MS) assessment of whether–and if so, when–13C glucose became detectable in the subjects’ blood is outlined in the next section.

Subjects were contacted approximately 1 day after study participation to assess for complications. Follow-up was achieved in all cases.

Tracer glucose infusate preparation and chemical analysis

The study design incorporated a form of tracer glucose, U-13C6 (the U represents ubiquitous, meaning in this case that all of the carbons in the glucose are 13C). This compound, not naturally found in the human body, is often used in metabolic studies [19,20]. The U-13C6 glucose used in the study was purchased locally (Cambridge Isotopes, Cambridge, MA) and combined in the MGH Research Pharmacy with standard glucose in a ratio of 1 g (of U-13C6) to 11.5 g (of standard glucose). The 12.5-g glucose, 1 g of which was U-13C6, was added to a 250-mL bag of NS to create an individual study subject’s infusate unit volume of 250 mL D5NS. Before infusion into study subjects, infusates were pumped through 0.2-um filters and tested for bacteriologic/pyrogen contamination.

Once blood samples were obtained, they were placed on ice before a short transport to the MIT laboratory. The process for GC/MS followed standard methodology [21]. In brief, 50 uL of plasma diluted with 1 mL cold MeOH was

Fig. 1 Placement of EASI access line. A, EASI access line placement–skin fold/preparation. B, EASI access catheter in SC space. C, Recombinant hyaluronidase injection preinfusion. D, EASI access line placement–connection of infusion.

deproteinized by 10 minutes’ centrifuging. Liquid super- natant was decanted and diluted with 2 mL hexane, then vortexed at high speed. After removal of the top (hexane) layer, the MeOH/sample layer was dried under nitrogen gas, with a heating block in a fume hood. Next, 150 uL butylboronic acid in pyridine was added; and the mixture was vortexed and heated at 95?C for 30 minutes. Upon cooling, 50 uL acetic anhydride was added, the mixture was vortexed, and the solution was allowed to stand for at least 30 minutes at Room temperature. After drying at 80?C and after addition of 0.5 mL hexane, the solution was transferred to microvials (for 1-uL injection) for analysis in a Hewlett Packard HP MSD-5973 (mass selective detector) and HP 6890 gas chromatographer with HP7683 injector (Hewlett Packard instruments by Agilent Technologies, Santa Clara, CA). Samples were run in electron ionization

mode using a capillary column, with helium as the carrier gas at a constant flow of 1 mL/min. The injection port temperature was 250?C; and the oven temperature began at 150?C, then (after 30 seconds) increased to 280?C at a rate of 30?C/min, before being held at 280?C for 4.8 minutes (total run time, 9.63 minutes).


Descriptive analysis was used to assess data such as subjeCT characteristics. For continuous and ordinal data, descriptive techniques included calculation of medians with IQR.

Access placement times were recorded as “number of 15-second increments” required for establishing access. For

example, if IV placement required 1 minute, the time for placement in 15-second intervals is “4.” This collapse of placement times from the continuous variable to an ordinal one was an a priori judgment made for a number of reasons. Practically speaking, it is not easy to reliably judge–particularly with multiple independent assessors doing the timing in different subjects–precisely when an IV is placed. From a clinical perspective, the investigators judged a few seconds’ difference in access times to be of marginal significance. Thus, the Information loss inherent in using 15-second intervals is offset by the 15-second segment method’s ease of execution; and the 15-second time frame constitutes sufficient resolution of placement times to allow meaningful conclusions about IV vs EASI access initiation.

Univariate analysis focused upon comparison of EASI and IVaccess for outcomes such as placement time, ease, and discomfort. Categorical variables were compared using Fisher exact test. For nonnormal continuous or ordinal (eg, Likert scale) data, nonparametric Kruskal-Wallis testing was performed.

The a priori focus of GC/MS data analysis was simply to determine if and when tracer glucose was detectable in the bloodstream. Because such detection occurred at the initial (t15) blood draw in essentially all cases, an a posteriori plan was made to use simple fixed-effect (study subject) regression to assess linearity of tracer uptake over the 60- minute blood sampling period.

Analysis was performed with STATA 10/MP (StataCorp, College Station, TX). Statistical significance was set at the P less than .05 level.


The 20 subjects in whom IV access and EASI access were placed ranged in age from 21 to 54 years, with a median of 43 years (IQR, 30-51). All were white, except for 2 female subjects (1 Pakistani and 1 Brazilian).

IV and EASI access establishment: placement time, ease, and subject discomfort

Catheter placements went easily. One subject underwent multiple attempts at IV cannulation, requiring 3 sticks. All EASI access catheter placements were successful on the first attempt. There was no difference (P = 1.0) in likelihood of first-attempt success for EASI (20/20, 100%) as compared with IV access (19/20, 95%).

In assessing times for IV and EASI access catheters, the unit of measurement was the number of 15-second intervals. All 20 EASI access catheters were placed within 15 seconds, whereas only 6 (30%) of IV catheter placements required less than 15 seconds (P b .001). In 9 cases (45%), IV catheter placement required at least a minute; and for 1 subject, 3

minutes elapsed before successful IV cannulation. When the times for IV placement were analyzed by the a priori plan of collapsing into 15-second intervals, nonparametric analysis found significantly (P b .0001) lower placement times for EASI access. The median number of 15-second intervals required for IV access establishment was 3, with an IQR of 1 to 6.5.

A 5-point Likert scale was used for operator-rated placement ease, with a score of 1 corresponding to the easiest rating and 5 to the most difficult. The EMTP operators assigned the easiest rating (1) to all 20 EASI access placements, as compared with 15/20 (75%) of IV placements (P = .047). The EMTPs assigned “2” ratings for IV placement to 3 subjects (15%), and 1 subject each received placement ease ratings of “3” and “4.” In categorical testing, exact analysis revealed an association between access type and EMTP-rated placement ease (P = .047). When results were analyzed ordinally, nonparametric testing confirmed the finding that EASI placement was rated significantly easier than IV placement (P = .019).

Subject-rated access placement discomfort was performed by comparing NRS pain scores (0-10). There was no difference (P = .672) between the NRS scores for EASI and IV access; for both groups, the median NRS was 1, with identical IQR of 0.5 to 2. In no case was the injection of HRH associated with subject-reported pain.

IV and EASI access fluid administration: infusion time and subject discomfort

The IV fluid was infused at a “wide-open” (maximal gravity assisted) rate in all subjects. For the 250-mL infusate, the median number of minutes for infusion was 20 (range, 15-35; IQR, 15-35). The mean infusion time was

22.5 minutes, with an SD of 7.3.

For the first subjects, the initial EASI access infusion was run at a wide-open rate (approximately 1000 mL/h). One of the first study subjects reported EASI infusion rate- associated discomfort. Some early subjects had no infu- sion-associated discomfort despite maximal gravity-assisted infusion rates, but the overall experience indicated that most subjects had an uncomfortable sensation associated with maximal gravity-assisted EASI infusion rates. Because the maximum achievable infusion rates were not a primary end point of the study, a decision was made by on-site physician investigators to eliminate the risk of subject discomfort by adjusting EASI infusion rates well below the maximal gravity-assisted rates to an operator-set rate of approximately 150 mL/h. There were no other instances of EASI infusion- associated discomfort after the rates were adjusted.

The artificial reduction of EASI access infusion times to well below the maximally allowable rates and the absence of confirmation that rates higher than those used would have resulted in significant pain limit the utility of the infusion times calculated for EASI access infusion. With that caveat, the median infusion time for administration of the 250-mL

EASI infusate was 125 minutes (range, 80-175; IQR, 95-155). The mean infusion time was 124.7 minutes (SD, 28.6). The EASI access infusion time was thus significantly longer (P b .001) than the IV access infusion time.

The median pain 1 minute into IV infusion was 0 (on a 10-point scale). The range was 0 to 2, with an IQR of 0 to 1. The median pain 1 hour after institution of IV infusion was 0; the range was 0 to 5, with an IQR of 0 to 1. As described above, the IV access line was removed in 1 subject because of discomfort.

The median pain 1 minute into EASI access infusion was 2 (on a 10-point scale); the range was 0 to 8 (with the “8” being in the above-mentioned subject who had pain at a wide-open EASI rate), with an IQR 1 to 3. This was significantly greater (P = .002) than the pain 1 minute into the IV infusion. The median pain 1 hour after institution of EASI access infusion was 0 (on a 10-point scale); the range was 0 to 3, with an IQR of 0 to 1. This was similar to the pain assessed at 60 minutes in the IV infusion group (P = 1.0).

Overall, when subjects were asked to rate their overall preference for access type (including placement and infusion), 8 preferred EASI and 12 preferred IV access. There was no association (P = .787) between operator and subject preference for IV vs EASI access. There were also no associations between preference for IV vs EASI access and age (P = .805), sex (P = .362), or whether EASI or IV access lines were placed in the dominant-hand upper extremity (P = 1.00).

Subjects with incomplete data

As this was a volunteer trial, with no benefits to the study subjects, the EASI access study protocol incorporated a low threshold for study dropout. In 2 cases, subjects’ EASI access catheters were removed before completion of the EASI access infusate; in 1 of these subjects, the IV was also removed for discomfort reasons. There was no statistically significant difference between rates of removal of EASI (2/ 20, 10%) and IV (1/20, 5%) catheters (P = 1.00).

One subject experienced “swelling and pressure” pain at the EASI access site within the first 5 minutes of SC infusion. Because of the a priori plan to immediately remove the EASI access line for any significant pain (as defined by the subject), EASI infusion was terminated and the EASI access catheter was removed. This subject also wished his IV access line removed (because of pain), and there were no phlebotomies performed other than the initial (t0) blood draw. This subject, who indicated preference for the IV route over EASI access, had Rapid resolution of pain at the EASI access site and was without sequelae during the 5-hour study period and upon follow- up. A second subject had pain self-rated as “not significant,” but the NRS of 5/10 prompted removal of the EASI access line. This subject actually indicated overall preference for the EASI access line. Her IV was left in place but was “positional,” and the only postinfu-

sion blood draw that was successful was at the t60 time (upon partial removal of the catheter).

In addition to the 2 subjects just discussed (one of whom had t0 phlebotomy only and one of whom had only t0 and t60 blood draws), there were 4 other subjects who did not have the full set of 5 phlebotomies. All of these subjects had a baseline (t0) phlebotomy with 3 (instead of the planned 4) additional blood draws. One subject each had failure of the blood draws at t15, t30, t45, and t60.

GC/MS results

Although there were cases for which there were less than the a priori-intended 5 blood samplings, all subjects except for 1 (the case in which both EASI and IV catheters were withdrawn) had at least 1 phlebotomy sampling performed after the t0 (baseline) sampling. The GC/MS results thus cover 19 subjects.

Fig. 2 depicts the plots for U-13C6 uptake over time. The GC/MS results revealed rapid (by t15) detection of U-13C6 in every case, although 1 subject (number 5 on the Fig. 2 plot) had minimal uptake.

Fixed-effect linear regression, adjusting for subject- related variability, was conducted on an a posteriori basis solely to provide hypothesis-generating information about the linearity of U-13C6 uptake. The subject variable accounted for 20% of the variation in the model, with an R2 of 0.76 (P b .001) revealing high correlation between time and U-13C6 uptake over the study period. The a posteriori nature of this analysis and the limited time frame over which the data were obtained precluded more in-depth analysis.

IV and EASI access site follow-up

There were no serious adverse effects or complications at either IV or EASI access sites. No physician reassess- ments were necessary for access site issues (or other reasons). Because any serious follow-up issues were to be evaluated with physician reassessment and because no such reassessments were necessary, nonserious issues were followed-up over the 1 to 3 days poststudy with at least 1 telephone call or e-mail (in a few cases, investigators incidentally encountered subjects at work). Thus, both the nature and the times of offset of infusion site issues are not known with precision. Access site issues that were encountered all resolved within a few days of study participation. Intravenous access site issues consisted of ecchymosis and tenderness reported in 11 of 20 subjects (55%); there were no such ecchymoses or tenderness at the EASI access sites (P b .001). All subjects in whom the full EASI access infusate volume (of 250 mL) was administered reported nonpainful swelling in the EASI access extremity; there was no swelling in the IV access extremities (P b

.001). The median extremity circumference increase noted

Fig. 2 Intravascular uptake of 13C-glucose into intravascular compartment.

at the conclusion of the EASI access infusion was 3 cm, with range 0 to 6 cm (IQR, 1.5-3.8).


There are many cases in the prehospital setting in which the desire for rapid access to the intravascular compartment outpaces the ability to quickly achieve such access. The problem may be a “tough stick” in an individual patient, or there may be dozens of MCI patients with only a few ALS providers. Intravenous access, with occasional IO backup, remains the standard of care; but it is easy to imagine the potential utility of an alternative approach.

Although there is some controversy about prehospital fluids administration [22], clinical practice and standard texts emphasize the importance of early fluid resuscitation [3]. Literature addressing MCI situations (eg, Crush injuries) makes a strong case for the importance of Fluid replacement [1]. Some of the recent literature on MCI has addressed the issue of delayed or unavailable IV access in both civilian and combat situations, with hypothesized solutions including such measures as proctoclysis–the infusion of fluids into the rectum [23]. Hypodermoclysis— a term for SC fluid infusion–has been in use for decades

and offers some obvious advantages to approaches such as proctoclysis [24,25].

The major historical reason for lack of use of hypoder- moclysis is the diffusion barrier presented by the interstitial matrix. Human recombinant hyaluronidase-facilitated EASI access aims to temporarily overcome this diffusion barrier. Previous studies have demonstrated the utility of EASI access and hypodermoclysis, but prior analyses have not included assessment of intravascular uptake of infusate. This trial drew upon previous metabolic study methodologies to incorporate assessment of uptake using tracer-labeled glucose and GC/MS [26]. The 60-minute time frame of our assay was better suited for its limited intent–to assess if and when tracer-labeled infusate was detected–than for any a posteriori analyses related to uptake rate.

The absorption data for the U-13C6 clearly demonstrate success in reaching the major goal for which these data were collected. The reliable detection of U-13C6 at the first phlebotomy sampling confirms the theory that EASI- administered infusate rapidly begins to reach the intravas- cular compartment. In 1 case (Fig. 2, subject number 5), both absorption and mole-ratio climb rate of U-13C6 were low. There were no obvious explanations for this result because the EASI access and the IV phlebotomy procedures both worked without problem in this subject. Future studies that follow tracer-labeled infusate for a longer period may be

necessary to shed more light on the explanation for, and the expected occurrence rate of, low absorption of EASI infusate. The results with respect to EASI access catheter placement confirm that, with minimal training, EMTPs can place the access lines with 100% success in a period significantly shorter than that required for IV access. The savings of a minute or two may not be clinically significant in all circumstances; but in the prehospital or MCI setting, even minimal Time Savings can be important. There is further reason for optimism in the finding that EMTPs were able to universally place EASI access on the first attempt and in less than 15 seconds. With EMTPs rating every EASI access placement at the level of “1” (easiest) on a 5-point difficulty scale, it seems worth considering (for future study) whether

EASI access placement in fact requires ALS-level training. This study’s identification of some problems with EASI highlights the need for further investigation. Most impor- tantly, some subjects’ infusion-associated pain and swelling stand in contrast to previous investigations in volunteers and patients [16,17]. In these previous studies, many types of fluids (including isotonic and hypertonic infusates) were given in greater volumes and at faster rates than those of the current trial [16,17]. Thus, consideration of mechanisms for EASI infusion’s adverse effects in the current study is useful to guide future investigation and clinical application

of the technique.

Other studies may have used more liberal (eg, as determined by clinical significance) definitions of pain and swelling. In the current study, any presence of pain or swelling, no matter how small, triggered end point definition. The investigators feel that this conservative approach is appropriate, but it is noteworthy that the levels of pain experienced with EASI infusion were both low and easily controllable (with reduction of infusion rate to about 150 mL/h).

Some other potential explanations for the pain and swelling findings in the current study were also considered. Previous studies used smaller catheters, but the infusion rate similarities render catheter size an unlikely explanation for differences in infusion-site findings. Because EMTPs placed every catheter quickly and easily, it also seems unlikely that there were operator differences accounting for subjects’ signs and symptoms. The potential differences explaining infusion pain and swelling are thus narrowed down to subject and infusion characteristics.

The study subjects were healthy and may have had different responses in the SC tissue than did patients in the palliative care setting [17]. However, other EASI study groups’ data denote painless (and swelling-free) use of lactated Ringer solution in volunteers [16]. Thus, although conclusions are necessarily based upon speculation, the investigators believe that the differences in infusion site signs and symptoms are related primarily to the infusate. Subcutaneous infusion of the 2 components (dextrose and saline) of the EASI access study is known to be nonproble- matic [17,24], so it appears that the tonicity of the

combination (rather than anything intrinsic to either of the 2 components) is responsible for the pain and swelling seen in the current study. There is physiologic plausibility to reasoning that, in this study’s healthy subjects, swelling is due to fluid flux into a hypertonic infusate deposit. The investigators believe this hypothesis should be tested by use of near-isotonic solutions such as D5W. This will also allow further evaluation of a potential limitation of EASI access: pain-limited infusion rates.

In addition to assessing varying infusate types and absorption rates, future EASI access investigations may focus on EASI administration of drugs such as analgesics [12,14]. In the hospice community, where SC fluids administration is not uncommon, EASI access has been used for administration of electrolytes, opioids, and other drugs [14,17]; and the ability to easily place an indwelling line for drug therapy may be attractive in some prehospital or MCI situations. Given the frequency with which EASI has been used for drug delivery–especially for opioids [25]— the possibilities for investigative assessment of its utility for this indication warrant further attention.

Overall, the EASI access study results are promising. As an easily learned, quickly performed, nonpainful means to gain access to the intravascular compartment, EASI warrants further attention for its possibilities in prehospital and acute care. Enzyme-assisted SC infusion may develop into a viable complement to IV access; and if the procedure proves reliable, it may in some cases be preferable to IO access– which is more difficult to perform and which has adverse effects of significant pain and possible bony injury. The most important contribution of the current study data may be the clear demonstration of rapid and consistent uptake of EASI- infused fluid into the intravascular space. Further studies should focus on infusion of different fluids in varying populations. Follow-up studies may include assessing feasibility of Basic Life Support provider placement of EASI access lines. Finally, additional investigation may address, in precise quantitative terms, intravascular uptake rates for various compounds.


The authors are grateful for the assistance from John Vetrano and the MGH Research Pharmacy, and from Jeff Breu and the MIT Clinical Research Center Core Labora- tory. The authors are also grateful to the study subjects, who were drawn largely from members of the MGH disaster response team.


  1. Sever MS, Vanholder R, Lameire N. Management of crush-related injuries after disasters. N Engl J Med 2006;354(10):1052-63.
  2. Stafford PW, Blinman TA, Nance ML. Practical points in evaluation and resuscitation of the injured child. Surg Clin North Am 2002;82(2): 273-301.
  3. Advanced trauma life support. 7th ed. Chicago: American College of Surgeons; 2004.
  4. Axelrod E, Strauss R, Kramer D. double blind comparative effectiveness of ELA-Max and EMLA in decreasing pain associated with adult venipuncture, protocol 99024–dataon file. Ferndale (Mich): Ferndale Laboratories, Inc; 2000.
  5. Riendeau LA, Bennett D, Black-Noller G, et al. Evaluation of the Analgesic efficacy of EMLA cream in volunteers with differing skin pigmentation undergoing venipuncture. Reg Anesth Pain Med 1999;24 (2):165-9.
  6. Rogers TL, Ostrow CL. The use of EMLA cream to decrease venipuncture pain in children. J Pediatr Nurs 2004;19(1):33-9.
  7. Soueid A, Richard B. ethyl chloride as a cryoanalgesic in pediatrics for venipuncture. Pediatr Emerg Care 2007;23(6):380-3.
  8. Taksande A, Vilhekar K, Jain M, et al. Pain response of neonates to venipuncture. Indian J Pediatr 2005;72:751-3.
  9. Zempsky WT, Sullivan J, Paulson DM, et al. Evaluation of a low-dose lidocaine iontophoresis system for Topical anesthesia in adults and children: a randomized, controlled trial. Clin Ther 2004;26(7):1110-9.
  10. Fowler R, Gallagher JV, Isaacs SM, et al. The role of intraosseous vascular access in the out-of-hospital environment (resource document to NAEMSP position statement). Prehosp Emerg Care 2007;11(1):63-6.
  11. Craig A, Eikenberry E, Parry D. Ultrastructural organization of the skin: classification on the basis of mechanical role. Connect Tissue Res 1987;16:213-23.
  12. Bookbinder L, Hofer A, Haller M, et al. A recombinant human enzyme for enhanced interstitial transport of therapeutics. J Control Release 2006;114(2):230-41.
  13. Laurent U, Dahl L, Reed R. Catabolism of hyaluronan in rabbit skin takes place locally, in lymph nodes, and liver. Exp Physiol 1991;76(5): 695-703.
  14. Frost GI. Recombinant human hyaluronidase (rHuPH20): an enabling platform for subcutaneous drug and fluid administration. Expert Opin Drug Deliv 2007;4(4):427-40.
  15. Baxter Healthcare Corporation. Hylenex package insert; 2006. Deerfield (Ill).
  16. Thomas JR, Yocum RC, Haller MF, et al. Assessing the role of human recombinant hyaluronidase in gravity-driven subcutaneous hydration: the INFUSE-LR study. J Palliat Med 2007;10(6):1312-20.
  17. Pirrello RD, Ting Chen C, Thomas SH. Initial experiences with subcutaneous recombinant human hyaluronidase. J Palliat Med 2007; 10(4):861-4.
  18. Thomas J, Yocum R, Haller M, et al. Hylenex recombinant hyaluronidase human injection dose-comparison study of sub- cutaneous hydration: the INFUSE-LR study; 2006 [abstract presented at Cleveland Clinic Palliative Medicine 2006, Sarasota FL, March 23-25].
  19. Wang P, Fraser H, Lloyd SG, et al. A comparison between ranolazine and CVT-4325, a novel inhibitor of fatty acid oxidation, on cardiac metabolism and left ventricular function in rat isolated perfused heart during ischemia and reperfusion. J Pharmacol Exp Ther 2007;321(1): 213-20.
  20. Wallis GA, Yeo SE, Blannin AK, et al. Dose-response effects of ingested carbohydrate on exercise metabolism in women. Med Sci Sports Exerc 2007;39(1):131-8.
  21. Beylot M, David F, Brunengraber H. Determination of the 13C- labeling pattern of glutamate by gas chromatography-mass spectro- metry. Anal Biochem 1993;212(2):532-6.
  22. Alam HB, Rhee P. New developments in fluid resuscitation. Surg Clin North Am 2007;87(1):55-72, vi.
  23. Kramer G, Hoskins S, Copper N, et al. Emerging advances in burn resuscitation. J Trauma 2007;62(6 Suppl):S71-2.
  24. Turner T, Cassano A. Subcutaneous dextrose for rehydration of elderly patients–an evidence-based review. BMC Geriatrics 2004;4 (2):1-6.
  25. Dalal S, Bruera E. Dehydration in cancer patients: to treat or not to treat. J Support Oncol 2004;2(6):467-479, 483.
  26. Kozak I, Kayikcioglu OR, Cheng L, et al. The effect of recombinant human hyaluronidase on dexamethasone penetration into the posterior segment of the eye after sub-Tenon’s injection. J Ocul Pharmacol Ther 2006;22(5):362-9.