Insulin Infusion Sets: A Critical Reappraisal

Abstract

An insulin infusion set (IIS) is a key component of insulin pumps. In daily practice issues with the IIS appear to be as relevant for a successful insulin therapy as the pumps themselves. The insulin is applied to the subcutaneous tissue via a Teflon^® (Dupont, Wilmington, DE) or steel cannula. There are intensive discussions about the impact the choice of material for insulin application has on insulin pharmacokinetics. In this review, this factor and others that are known to have an impact on the successful usage of IIS are discussed.

Background

An insulin infusion set (IIS) is required to transfer the insulin from the reservoir in the pump to the subcutaneous tissue with all types of insulin pumps. Insulin is infused via a steel cannula or via a Teflon^® (Dupont, Wilmington, DE) cannula at the tip of the IIS. For Teflon cannulas, a steel guiding needle can be retracted immediately after insertion into the skin, keeping the Teflon cannula in place, or the stainless steel cannula remains in the subcutaneous tissue for the wear-time of a given IIS. The focus of this review is on the following:

• Impact of the cannula material (Teflon vs. steel) on insulin pharmacokinetics (PK)/pharmacodynamics (PD) (i.e., on the insulin absorption at the infusion site)

• Impact of the depth of the infusion (determined by cannula length) on insulin PK/PD

• Pain associated with cannula diameter during insertion

• Other factors like ethnic differences in skin thickness.

The number of scientific publications about IISs is relatively small, which is in contrast to the high number of insulin pump users worldwide (>500,000).¹ In addition, the quality of many studies (when it comes to study design, sample size, study duration/performance) is mediocre. The paucity of data reported from manufacturers of insulin pumps, as they get a lot of reports from customers of complaints about issues with their pumps and IISs, is also a matter of concern for the European and U.S. diabetes medical associations (European Association for the Study of Diabetes and the American Diabetes Association, respectively).^2,3 It is of interest to note that none of the guidelines for insulin pump therapy provides clear recommendations pro or con for steel versus Teflon cannulas, nor does a detailed white paper of the American Association of Diabetes Educators (www.diabeteseducator.org/docs/default-source/legacy-docs/_resources/pdf/research/12-30-11-aade_insulin_whitepaper_print.pdf?sfvrsn=2).

Steel Versus Teflon—General Aspects

There is limited publicly assessable knowledge about how many pump wearers in different countries use steel versus Teflon cannulas. Survey data suggest that steel cannulas are more commonly used in Germany (approximately 40%) and Europe (approximately 25%) than in the United States (approximately 10%), as well as suggesting that most IISs nowadays use a Teflon cannula.⁴ This may correlate with the market share for various brands of insulin pumps in these countries. Within a given country, diabetologists/pump centers might have different views on the cannula material topic; this might also have an impact on the usage of different IISs. It appears as if the main reasons for the switch from steel to Teflon cannulas were business or patent aspects.

In contrast to the scarcity of scientific publications about the impact of the cannula material on insulin PK/PD, there is an intensive discussion about this topic in the Internet on patient blogs and lay journals. Patients report about significant differences in metabolic control depending on the cannula material they are using. Most often steel is mentioned to be better, also with respect to the variability of insulin action.

One explanation for the patient observation might be that at least in some patients the local inflammatory reaction (which has an impact on insulin absorption) is lower with steel than with Teflon.^5,6 In an oral presentation at a Food and Drug Administration Artificial Pancreas Workshop in 2013, Jeff Joseph from Philadelphia presented animal data (from pigs) indicating that Teflon cannulas induce more histological changes in the subcutaneous tissue than steel cannulas. However, these data were never fully published, and no respective studies exist with skin biopsies from humans. A better biocompatibility of steel than of Teflon cannulas was also reported by others.⁶

One reason for the failure of IISs is occlusions (i.e., no insulin is pumped through the tubing into the subcutaneous tissue). It appears as if the occurrence of occlusions is a complex multifactorial process, and clearly wear-time is one of these factors. Longer duration of an IIS increases the risk of building fibrils, which is one reason for an increase in occlusions over time.

Wear-Time

Length of use recommendations for IISs is generally 2 days for a steel IIS and 3 days for a Teflon IIS, although it is difficult to find sound data supporting for these suggestions. Many patients report that, in practice, they often use their IIS for several days or even longer without any issues. Several studies (see below) suggest that an IIS should be changed in 3-day intervals; however, as stated there is no clear proof that steel cannulas have to be changed more frequently than Teflon cannulas. One reason for this ambiguity is considerable differences between patients; these interindividual differences require large sample sizes in studies in order to enable observation of rare cases and/or make relative small patient groups significant.

Our knowledge about the wear-time of IISs that patients use in daily life is limited. It appears that some patients have to change them quite often (= daily), whereas others can use them for weeks without any skin reactions/worsening of glucose control induced by changes in insulin PK/PD. It would be interesting to have better data about the usage/wear-time of given cannula types in different countries and the frequency of adverse event reporting.

Wear-Time for Steel Versus Teflon

Several studies have evaluated the frequency of IIS failures and other adverse events depending on the type of IIS used. If the wear-time is different between steel/Teflon cannulas, this can have an impact on the number of adverse events observed. Some data indicate that the wear-time of steel cannulas in daily practice is shorter than that of Teflon cannulas.^7,8

A recent survey reported a mean duration of IIS usage of 3.2 days (range, 2–6 days).⁹ During an extended wear study of an IIS with a Teflon cannula and an IIS with a steel cannula for 7 days or until failure, the IIS with the Teflon cannula was found to have a 15% initial failure rate due to kinking on insertion.¹⁰ After the end of 1 week, the failure rate for steel and Teflon cannulas gradually equaled out at 64% for each IIS. The main predictor for use of IIS wear was the individual patient, not the type of IIS. During the entire 77 weeks of wear, 30% of IISs failed because of hyperglycemia following a failed correction dose, 13% were removed for pain, 10% were pulled out by accident, 10% had erythema and/or induration of >10 mm, 5% fell out because of loss of adhesion, and 4% were removed for infection.¹⁰ A comment on this study discussed whether the IIS with Teflon cannula studied in this study has a higher risk of failures than others (i.e., that not all Teflon cannulas should be considered as being similar).¹¹

In a survey with 432 pediatric patients with type 1 diabetes, 56% used IISs with Teflon cannulas, and 44% used ones with steel cannulas.¹² No differences in the number of complications were observed: 40% showed up on the first day of usage, with 83% by the end of the second day of usage.

Kerr et al.¹³ demonstrated that early IIS occlusions (within 3 days) are rare and independent of the type of rapid-acting insulin analog studied. These authors have recommended that the IIS should be replaced within 3 days irrespective of the insulin used.

Another study compared 45 pump wearers' standard IISs (44 Teflon and one steel) worn over a 1-month period with an IIS with a Teflon cannula (Accu-Chek^® FlexLink infusion set; Roche Diagnostics, Mannheim, Germany) studied during its third month of use in the trial. Insertion devices were used by 80% of participants with their standard IIS and by 93% with the IIS being tested. Set failure, defined as immediate or unexplained hyperglycemia within the next 6 h that could not be corrected, occurred during 8.9% of insertions of the standard IIS (45 of 507 insertions). With the Accu-Chek FlexLink, the initial failure rate fell to 3.2% (15 of 488 insertions).¹⁴ Initial failures were surprisingly common and were usually associated with the use of an auto-insertion device. They occurred on the standard IIS once every 22.5 days with 2 days of wear and once every 33.7 days with 3 days of wear. Subsequent to these initial failures, an additional 8% of IISs failed in both groups. Just over 400 participants (42%) had to replace an IIS early at least once during this study in each of the IIS test groups. One participant experienced diabetic ketoacidosis, and six had measurable levels of ketones with hyperglycemia during the 60 days of study. Despite the promising results of this research study, the FlexLink Plus studied was recalled in February 2011 because of excessive kinking and bending of the cannula. It is worth mentioning that the manufacturer showed responsibility and redesigned the ISS. The new version was launched 2014 together with the Accu-Chek Insight pump system as the Accu-Chek Insight Flex system.

Patient Comfort

Another aspect in which steel and Teflon cannulas may differ is patient comfort. Many patients and physicians assume that the comfort is lower with a steel cannula (= more rigid) than with a flexible Teflon cannula, especially during physical work or sports activities. However, it appears that in reality this is not the case. It is of interest to note that it appears as if this question was never formally addressed in a larger number of patients by means of structured questionnaires (= no scientific publication); however, patients report in their blogs that steel cannulas are simpler and easier to insert and as comfortable as Teflon cannulas, and statements about differences with respect to comfort can also be found in publications.¹⁵ In an evaluation of 2,452 new pump users in Germany, 177 switched to other forms of insulin therapy; younger age by 12 months and use of Teflon cannulas were related to lower odds of pump discontinuation.¹⁶ However, in this study 80% of the patients (most of these had an age of <18 years) used an IIS with a Teflon cannula, which is different from adult patients in Germany (see above).

Patient comfort might be different depending on how the cannula is applied and if the insertion is done perpendicular or with a given angle. There might be also differences within a group of IISs (i.e., the question is if all Teflon cannulas are the same when it comes to comfort). Compared with steel cannulas, where an insertion device is not needed, auto-insertion devices are commonly used with Teflon IISs. For most dexterous individuals, manually inserting a steel cannula provides better control with elimination of kinking and a lower rate of insertion failure compared with use of an insertion device.

It is not clear if steel cannulas made by different manufacturers are truly identical in their performance; the surface coating, polishing, the material itself, etc., might vary to a given extent among them. The same holds true for Teflon cannulas. Depending on, for example, the rigidity of the cannula, the risk of bending/kinking of the cannula might vary.¹¹

Wear-Time and Glucose Control

In a study from 1991 (with different insulin, pumps, IISs, etc.) a bolus of regular human insulin was infused in 15 patients with type 1 diabetes.¹⁷ Peak free plasma insulin levels were observed earlier on the fourth day of using the IIS in comparison with the first day (56 vs. 110 min; P < 0.01). This was the first description of a change in insulin PK/PD over the wear-time of an IIS. However, in a study from 1993 with subcutaneous infusion of regular insulin for 4 days no significant differences in levels of blood glucose and plasma free insulin could be observed over time.¹⁸ The relevance of such old studies for the pump therapy of today is doubtful as practically all materials have changed a lot in the last 25 years, and in these studies the factor of cannula material was not taken into account.

In a more recent randomized study comparing two different insulin analogs (insulin lispro vs. insulin aspart) the patients used the IIS for 5 days.¹⁹ The researchers reported increased glucose levels from Day 3 to 5 (from 144 to 164 mg/dL, despite an increase in total insulin dose from 49 to 55 IU/day). Therefore they recommend change of the IIS every 2 days. In another study with 12 subjects the individuals were asked to wear an IIS with a Teflon cannula for up to 5 days; the authors reported that the mean daily glucose levels increased slightly but not significantly from a mean of 135 mg/dL on Day 1 to 162 mg/dL on Day 5.²⁰

On one hand, there is at least a tendency to a worsening of metabolic control (= increase in mean glucose levels over the days wearing the same IIS); on the other hand, it is known that insulin absorption is improved at the infusion site.²¹ It is of note that it was not clearly stated in the publication which type of IIS the patients used (“their own insulin pump and type of infusion set”); however, as the study²¹ was performed in the United States, it can be safely assumed that most if not all of the IIS had a Teflon cannula. The better insulin absorption after some days of wear-time is most probably related to an increase in local blood flow.^6,22 Probably the prandial insulin requirements are covered more optimally with an increase in wear-time, but the insulin infused with the basal infusion rate (which can be assumed to remain constant over the days) induces a reduced metabolic effect.

In a multicenter, 13-week study with 256 subjects wearing an IIS with a Teflon cannula for a mean duration of 71 h, unexpected hyperglycemia and/or IIS set occlusion occurred in 61–68% of the subjects depending on which rapid-acting insulin analog they used.¹⁹ An Internet survey organized by the T1D Exchange clinic network showed that the 243 participating adults use their IIS for 3 days and that the mean fasting blood glucose level increases with each successive day of IIS use: from 126 mg/dL on Day 1 to 133 mg/dL on Day 3 to 147 mg/dL on Day 5.²³ Most patients used an IIS with a Teflon cannula (57%), and only 4% reported using a steel cannula; however, 40% made no statement. Recent data from a prospective randomized controlled crossover study with 2 × 3-month observation periods with 22 patients with type 1 diabetes using an IIS with a Teflon cannula for 2 or 4 days showed significantly more IIS-related adverse events with the longer wear-time (290 vs. 495; P < 0.05) but no difference in metabolic control.²⁴ In conclusion, in daily life most data indicate that longer usage of IIS is associated with worsening of glucose control.

Steel Versus Teflon: Insulin PK/PD

There is only one study in which the metabolic effect of using two different types of IISs and the wear-time of the IIS was evaluated (see above).¹⁰ By means of a continuous glucose monitoring system, glucose control was followed during the wear-time of the IIS until hyperglycemia (>250 mg/dL) could no longer be controlled by a correction bolus. No significant increase in glycemia was observed over time, and also no difference in glucose control was observed between the two types of IIS.

In a head-to-head study between a patch pump (OmniPod^® insulin pump system; Insulet Corp, Bedford, MA) and a conventional pump (Paradigm^® Veo insulin pump system with the MMT-399 Quick-set Paradigm IIS with a tubing length of 60 cm; Medtronic Diabetes, Northridge, CA), the impact of wear-time on insulin PK was studied in 20 patients with type 1 diabetes, as well as the relevance of this on postprandial glycemic excursions.²⁵ With both types of pumps plasma insulin profiles were similar when applying the same (individualized) insulin dose on Day 1. However, on Day 3 the increase in insulin levels was more rapid, and higher maximal levels were achieved earlier with both pumps. This indicates that insulin PK is comparable with both types of insulin pumps, at least in this patient groups and types of pumps studied. Due to the fact that the speeds with which insulin pumps or the patch pump OmniPod deliver the same insulin dose are remarkable different, such evaluations have to made for each type of patch pump (when they come to the market) as the technology used for insulin infusion is quite different.

Real-World Experience with Steel Versus Teflon Cannulas with Respect to Glucose Control

Data from a recent survey with 1,142 patients in Germany confirmed that success of the IIS is highly relevant for daily success with pump therapy.⁸ The metabolic control of patients using an IIS with a steel cannula was not different from that of those using a Teflon cannula (glycosylated hemoglobin level of 7.7% vs. 7.8%); however, a smaller number of patients observed unexplained increases in glycemia with steel cannulas (in total, n = 405: yes, n = 210 [52%]; no, n = 195 [48%]) than with Teflon cannulas (in total, n = 535: yes, n = 313 [59%]; no, 222 [41%]) (P < 0.05 by χ² test).

Air Bubbles

After a “cold” cartridge (the insulin comes from the refrigerator) is put into the pump, this gets warmer when the pump is close to the body.²⁶ The gas inside the insulin solution forms air bubbles. These can enter the IIS and are transferred into the subcutaneous adipose tissue. By rule of thumb an air bubble with a length of 10 cm in the IIS accounts for 1 U of insulin; in other words, a totally empty 100–110-cm-long IIS equals approximately 12–13 U of insulin. That means that smaller bubbles (<2 cm) most probably have no clinical relevance in “normal” pump patients (with the exception of preschool kids). This clearly depends critically on the basal infusion rate (i.e., with low infusion rates insufficient amounts of insulin are applied, which might lead to hyperglycemia/increase in ketone bodies in blood). This is why occurrence of air bubbles should be avoided as much as possible (there are several respective instructions available in the Internet) and why other factors that might contribute to the occurrence of bubbles (like higher temperatures in the surrounding or higher altitudes) have to be taken into account.²⁷ There are reports that changes in altitude (while flying) might induce unintended insulin delivery from pumps.²⁸ However, such evaluations did not take the back pressure of the tissue into account (see below).

Pain Associated with IIS Insertion/Usage

When it comes to pain studies, most often the pain associated with IIS insertion is addressed. Therefore, not many data about the pain during usage of IIS are available. In a recent study two IISs with soft cannulas (= Teflon) were compared.²⁹ In an open-label, randomized, crossover, multicenter study, in total, 80 patients (20–74 years old; mean age, 46.5 ± 12.9 years; 58.8% male) were enrolled to evaluate the usage of two IISs (Accu-Chek FlexLink Plus and Accu-Chek FlexLink) under real-life conditions at home for a 4-week period each. All patients had to record details about every IIS insertion in a patient diary. Insertion-related pain was assessed by means of a visual analog scale. The mean pain noted during insertion was similar for both IISs: visual analog scale score, 6.1 ± 6.4 mm for the FlexLink Plus and 5.8 ± 6.3 mm for the FlexLink (P = 0.92). Twenty percent of the IIS changes with the FlexLink Plus and 16.0% with the FlexLink were unplanned. During the study 13 patients had adverse events, and two serious adverse events occurred. Of the patients, 18.7% had kinked cannulas. Most patients were satisfied with both IISs. Both IISs were safe and effective. The two IISs did not differ from each other with respect to pain observed during insertion.

The pain associated with cannula insertion might also depend on the angle with which the cannula is inserted: perpendicular or with a given angle; however, it appears as if no systematic study of this aspect was performed.

Cannula Length/Diameter and Skin Type/Ethnic Differences

Different insertion depths of a cannula (for example, injection of insulin with cannulas of different length) will apply the insulin into different layers of the subcutaneous tissue. There is a long discussion about the relevance of the insertion depth on insulin PK/PD without a clear-cut outcome.³⁰ For many years there was the belief that considerable differences exist between different ethnic groups, male/female, younger/older people, etc., with respect to skin thickness.

A systematic evaluation of the skin thickness of 388 U.S. adults with diabetes (in three subgroups by body mass index of <25, 25–29.9, and >30 kg/m²) revealed that the thickness at typical injection sites does not differ by clinically significant degrees in demographically diverse adults.³¹ Thus, the concern that a relatively thin steel cannula might bend when inserted into someone with a ”thick“ skin is most probably not realistic. In the same study also the thickness of the subcutaneous adipose layer was evaluated. As to be expected, the differences in the thickness of the fatty tissue vary largely depending on the body weight, etc., as well as depending on the gender. The total skin thickness (i.e., the skin itself) was in all patient groups <5 mm. The risk of an intradermal application of drugs like insulin is quite small, even with short 5-mm cannulas. Therefore, with infusion cannulas with 5–8 mm in length a strict subcutaneous infusion can be achieved.

Ultrasound measurement have determined that 6% of all 6-mm 90° injections will be intramuscular in adults.³¹ These measurements also included arms and thighs, which on average had 2–6 mm less subcutaneous tissue compared with the abdomen or buttocks. In a pediatric study the mean thickness of skin and subcutaneous tissue at the arm was 6.3 mm, that over the thigh was 7.5 mm, that over the abdomen was 8.0 mm, and that over the buttocks was 8.1 mm.³²

In a study with six patients with type 1 diabetes three different IISs with Teflon cannulas were studied.³³ The cannula length of one IIS was 6 mm, and the other two had a length of 9 mm. In this study blood ketone bodies were measured during wear-time periods of 4 days. The cannula length had no impact on ketone levels.

Pain perception associated with insertion of cannula into the skin depends on a variety of factors. In such a multifactorial situation it is not easy to focus on one factor (e.g., cannula diameter) only with respect to the pain induced. Many of the studies performed have methodological flaws (e.g., the experimental design was not “blinded,” and the patients could see when the insertion was performed).

The outer and inner diameters of Teflon cannulas are larger than those of steel cannulas. The typical outer diameter of steel cannulas is 26–29 gauge; with thinner cannulas the pain perception is usually lower.³⁴ However, one reason why insertion of Teflon cannulas can be more painful (and why auto-inserters are used) is the steel needle that is needed to insert the Teflon cannula, which clearly has to have an even larger diameter. Nevertheless, the pain associated with insertion of the steel cannulas depends critically on several other factors like how the tip of the cannula is formed, if the surface of the steel cannula is coated, etc.

In view of the lateral movement of the cannula while it is being used, induced by the tubing and body movement (which induces kinking/bending of Teflon cannulas), there is a risk of breaking the steel cannula off from its carrier. A similar risk of disconnection applies to Teflon cannulas. This would mean that a quite thin cannula remains in the skin or subcutaneous tissue.^35
–37 Although steel cannulas are easily detectable via X-rays, Teflon cannulas need absorbing additives to become “visible” to X-rays. Despite the general knowledge of the dislodged cannula's position, it may be difficult to localize and remove during surgery. Also, during the insertion process such a risk might exist if the skin is “hard,” and the user does not guide the cannula adequately.

Insertion

Clearly the way the insertion of the IIS is handled has a major impact on the pain associated with insertion of the cannula, insertion depths, etc. Again it appears as if some patients are using auto-inserters regularly (mainly with Teflon cannulas), whereas many other prefer manual insertion. No systematic evaluation was published if usage of an auto-inserter is associated with a better “outcome” of pump therapy. With the OmniPod the insertion of the infusion cannulas is usually done “automatically” by the patch. The technology used to do this varies widely among patch pumps (only the OmniPod is currently on the market, but several are supposed the enter the market soon), which in turn could mean that the pain perception associated (if there is any) differs. Insertion of a Teflon cannula is associated with a certain risk of bending or kinking the cannula during the insertion process or during the wear-time. Also, the angle in which the cannula is inserted can have an impact of the “success” of the insulin infusion: a perpendicular insertion might be different with respect to the risk of bending/kinking from one that is inserted slanted with a given angle. This can lead to an occlusion of the infusion line/blockage of insulin infusion. It is of note that the time required until an occlusion alarm is given by the insulin pumps differs widely as well, depending critically on the basal infusion rate. If this is low, it can be several hours before an alarm is provided.

Fixation

One issue is that the fixation of the cannula has an impact on the infusion site/insulin PK. In other words, if the catheter is not well attached to the skin and is pulling on the cannula more or less constantly, this has an impact on the wear-time/reliability of infusion site (= higher risk of leaks). As the situation between a conventional pump and a patch pump differs massively in this respect, one has to be careful with the transferability of data from results obtained with conventional pumps/IISs to patch pumps. It is difficult to say how “stable” the infusion cannula (Teflon in case of the OmniPod) sticks into the skin/subcutaneous tissue. This might vary among different patch pumps, depending on the area with adhesive material, height/weight of the patch, etc. Again, systematic studies by an independent site are missing.

The infusion line of the IIS should be anchored with tape, so that any tug or movement of the infusion line is not directly transferred to the cannula under the skin. If the fixation is not optimal, this can lead to skin irritation, redness, and on occasion leakage of insulin along the cannula back to the skin.

Local Skin Reactions/Lipodystrophy

It is not clear if the usage of steel cannulas for insulin infusion (with repeated insertion of the cannulas in the same skin area and prolonged usage) leads to a higher incidence of local skin reactions (= lipoatrophy or lipohypertrophy) than usage of Teflon cannulas. In a recent evaluation of dermatological side effects of pump therapy in children and adolescents, more patients using a Teflon cannula experienced any dermatological side effect than those using steel cannulas; however, no association between the frequency of skin alterations and the material of the cannula was found.³⁸ Twenty percent of the patients had lipohypertrophy, and 6% had lipoatrophy. Frequent occurrence of lipohypertrophy (42%) was also observed in other dermatological studies with 50 young patients with type 1 diabetes³⁹ and a survey with 92 adult patients (26.1%).⁹ In another with 78 children on insulin pump therapy, 46% of these reported lipohypertrophic areas at the insertion site.⁴⁰

The risk of skin reactions might also be influenced by which rapid-acting insulin analog (or human regular insulin) is infused. It is not the insulin per se that might have an impact, but the excipients in the insulin formulation.

Novel IIS designs

Recently data were published about significant, intermittent, and unexplained elevations in blood glucose levels in pump wearers that are associated with rapid increases in the pressure measured in the IIS used.⁴¹ Pressure sensors had been installed in these infusion lines. Because an occlusion alarm was never triggered and the hyperglycemia was able to be corrected over time, these episodes were described as “silent or hidden occlusions” and may explain some instances of unexplained hyperglycemia. The sponsor of that study will bring a new IIS to the market in the near future that is supposed to reduce the impact of such pressure increases.

In another recent study the resistance that the tissue built up against insulin infusion was monitored.⁴² Infusion of a liquid test solution via IIS with an 8-mm steel cannula with different infusion rates in 12 patients with type 1 diabetes and 12 healthy subjects showed an increase in this pressure with higher infusion rates. One consequence of this observation is that the sensitivity of the occlusion alarm should be adjusted to the infusion rate used.

Further improvements in CSII therapy with regard to IIS could be gained by longer wear-time intervals and easier detection of problems related to insufficient insulin delivery. Improvements could also be achieved through standardized patient training and systematic follow-up of problems after CSII initiation.

Conclusions

Based on the currently available evidence from published studies, no clear statement can be made pro or con for a steel versus a Teflon cannula when it comes to insulin PK/PD. However, this is mainly due to the paucity of studies, and practical experience of many diabetes nurses/educators indicates that steel cannulas are more reliable in daily practice of patients with diabetes on insulin pump therapy. Cannula length (= depth of infusion), pain, and ethnic differences also appear to have no relevant impact on the selection of the material used for the IIS. One would assume that it is relatively easy to evaluate this topic systematically under controlled conditions; however, it appears as if no such studies were initiated so far.

Footnotes

Acknowledgments

The creation of the manuscript was supported by an unrestricted educational grant from Roche Diagnostics GmbH, Mannheim, Germany.

Author Disclosure Statement

L.H. holds shares in the Profil Institute for Metabolic Research, Neuss, Germany, and the Profil Institute for Clinical Research, San Diego, CA. He is a consultant for a range of companies that develop new diagnostic and therapeutic options for the treatment of diabetes.

References

Heinemann

, Krinelke

: Insulin infusion set: the Achilles heel of continuous subcutaneous insulin infusion. J Diabetes Sci Technol, 2012; 6:954–964.

Heinemann

, Fleming

, Petrie

, Holl

, Bergenstal

, Peters

: Insulin pump risks and benefits: a clinical appraisal of pump safety standards, adverse event reporting, and research needs: a joint statement of the European Association for the Study of Diabetes and the American Diabetes Association Diabetes Technology Working Group. Diabetes Care, 2015; 38:716–722.

Heinemann

, Fleming

, Petrie

, Holl

, Bergenstal

, Peters

: Insulin pump risks and benefits: a clinical appraisal of pump safety standards, adverse event reporting and research needs. A joint statement of the European Association for the Study of Diabetes and the American Diabetes Association Diabetes Technology Working Group. Diabetologia, 2015; 58:862–870.

Walsh

, Roberts

, Weber

, Faber-Heinemann

, Heinemann

: Insulin pump and CGM usage in the United States and Germany: results of a real-world survey with 985 subjects. J Diabetes Sci Technol, 2015; 9:1103–1110.

Shetty

, Wolpert

: Insulin pump use in adults with type 1 diabetes—practical issues. Diabetes Technol Ther, 2010; 12(Suppl 1):S11–S16.

Hojbjerre

, Skov-Jensen

, Kaastrup

, Pedersen

, Stallknecht

: Effect of steel and Teflon infusion catheters on subcutaneous adipose tissue blood flow and infusion counter pressure in humans. Diabetes Technol Ther, 2009; 11:301–306.

Johansson

, Adamson

, Lins

, Wredling

: Patient management of long-term continuous subcutaneous insulin infusion. J Adv Nurs, 2005; 51:112–118.

Reichert

, Weber

, Kaltheuner

, Scheper

, Faber-Heinemann

, Heinemann

: Realität der Insulinpumpentherapie in Diabetesschwerpunktpraxen: Daten von 1142 Patienten aus 40 diabetologischen Schwerpunktpraxen. Diabetes Stoffwechsel Herz, 2013; 22:367–375.

Pickup

, Yemane

, Brackenridge

, Pender

: Nonmetabolic complications of continuous subcutaneous insulin infusion: a patient survey. Diabetes Technol Ther, 2014; 16:145–149.

10.

Patel

, Benasi

, Ferrari

, Evans

, Shanmugham

, Wilson

, Buckingham

: Randomized trial of infusion set function: steel versus Teflon. Diabetes Technol Ther, 2014; 16:15–19.

11.

Renard

, Boizel

: Comments on Patel et al.'s “Randomized trial of infusion set function: steel versus Teflon”: early catheter failures may differ according to Teflon catheter model. Diabetes Technol Ther, 2014; 16:545–546.

12.

Liebner

, Holl

, Heidtmann

, Lilienthal

, Molz

, Hungele

, Bergheim

, Gialeli

. Insulinpumpenkatheter Komplikationen im Kindes- und Jugendalter [abstract]. Diabetol Stoffwechsel, 2011; 6:S52.

13.

Kerr

, Morton

, Whately-Smith

, Everett

, Begley

: Laboratory-based non-clinical comparison of occlusion rates using three rapid-acting insulin analogs in continuous subcutaneous insulin infusion catheters using low flow rates. J Diabetes Sci Technol, 2008; 2:450–455.

14.

Renard

, Guerci

, Leguerrier

, Boizel

: Lower rate of initial failures and reduced occurrence of adverse events with a new catheter model for continuous subcutaneous insulin infusion: prospective, two-period, observational, multicenter study. Diabetes Technol Ther, 2010; 12:769–773.

15.

Rice

, Sweeney

: Choosing and using an insulin pump infusion set. Diabetes Self Manag, 2006; 23:60, 62–64, 67.

16.

Kostev

, Rockel

, Rosenbauer

, Rathmann

: Risk factors for discontinuation of insulin pump therapy in pediatric and young adult patients. Prim Care Diabetes, 2014; 8:346–351.

17.

Liu

, Moberg

, Wredling

, Lins

, Adamson

: Insulin absorption is faster when keeping the infusion site in use for three days during continuous subcutaneous insulin infusion. Diabetes Res Clin Pract, 1991; 12:19–24.

18.

Olsson

, Arnqvist

, Asplund

: No pharmacokinetic effect of retaining the infusion site up to four days during continuous subcutaneous insulin infusion therapy. Diabet Med, 1993; 10:477–480.

19.

van Bon

, Bode

, Sert-Langeron

, DeVries

, Charpentier

: Insulin glulisine compared to insulin aspart and to insulin lispro administered by continuous subcutaneous insulin infusion in patients with type 1 diabetes: a randomized controlled trial. Diabetes Technol Ther, 2011; 13:607–614.

20.

Schmid

, Hohberg

, Borchert

, Forst

, Pfutzner

: Pilot study for assessment of optimal frequency for changing catheters in insulin pump therapy—trouble starts on day 3. J Diabetes Sci Technol, 2010; 4:976–982.

21.

Swan

, Dziura

, Steil

, Voskanyan

, Sikes

, Steffen

, Martin

, Tamborlane

, Weinzimer

: Effect of age of infusion site and type of rapid-acting analog on pharmacodynamic parameters of insulin boluses in youth with type 1 diabetes receiving insulin pump therapy. Diabetes Care, 2009; 32:240–244.

22.

Clausen

, Kaastrup

, Stallknecht

: Effect of insulin catheter wear-time on subcutaneous adipose tissue blood flow and insulin absorption in humans. Diabetes Technol Ther, 2009; 11:575–580.

23.

Sampson Perrin

, Guzzetta

, Miller

, Foster

, Lee

, Block

, Beck

: A web-based study of the relationship of duration of insulin pump infusion set use and fasting blood glucose level in adults with type 1 diabetes. Diabetes Technol Ther, 2015; 17:307–310.

24.

Pfützner

, Sachsenheimer

, Grenningloh

, Heschel

, Walter-Johannesen

, Gharabli

, Klonoff

: Using insulin infusion sets in CSII for longer than the recommended usage time leads to a high risk for adverse events—results from a prospective randomized crossover study. J Diabetes Sci Technol, 2015; 9:1292–1298.

25.

Luijf

, Arnolds

, Avogaro

, Benesch

, Bruttomesso

, Farret

, Heinemann

, Place

, Renard

, Scotton

, DeVries

: Patch pump versus conventional pump: postprandial glycemic excursions and the influence of wear time. Diabetes Technol Ther, 2013; 15:575–579.

26.

Herr

, Keith

, Klug

, Pettis

: Characterizing normal-use temperature conditions of pumped insulin. J Diabetes Sci Technol, 2014; 8:850–854.

27.

Lopez

, King

, Goss

, Chockalingam

: Bubble formation occurs in insulin pumps in response to changes in ambient temperature and atmospheric pressure but not as a result of vibration. BMJ Open Diabetes Res Care, 2014; 2:e000036.

28.

King

, Goss

, Paterson

, Crock

, Anderson

: Changes in altitude cause unintended insulin delivery from insulin pumps: mechanisms and implications. Diabetes Care, 2011; 34:1932–1933.

29.

Freckmann

, Amstutz

, Pickering

, Künsting

: Accu-Chek FlexLink Plus and its predecessor device cause low pain preceptions in a clinical randomized crossover trial [abstract]. Diabetes Technol Ther, 2015; 17(Suppl 1):A-60.

30.

Frid

, Linde

: Intraregional differences in the absorption of unmodified insulin from the abdominal wall. Diabet Med, 1992; 9:236–239.

31.

Gibney

, Arce

, Byron

, Hirsch

: Skin and subcutaneous adipose layer thickness in adults with diabetes at sites used for insulin injections: implications for needle length recommendations. Curr Med Res Opin, 2010; 26:1519–1530.

32.

, Ingegnosi

, Strauss

: Skin and subcutaneous thickness at injecting sites in children with diabetes: ultrasound findings and recommendations for giving injection. Pediatr Diabetes, 2012; 13:525–533.

33.

Ampudia Blasco

, Parramon

, Gabaldon

, Sanz

, Catala

, Carmena

: Evaluación de la seguridad de los catéteres de Teflón en pacientes diabéticos tipo 1 tratados con infusión subcutánea continua de insulina (ISCI) mediante la determinación de la cetonemia capilar. Endocrinol Nutr, 2006; 53:82–86.

34.

Hirsch

, Gibney

, Albanese

, Qu

, Kassler-Taub

, Klaff

, Bailey

: Comparative glycemic control, safety and patient ratings for a new 4 mm × 32G insulin pen needle in adults with diabetes. Curr Med Res Opin, 2010; 26:1531–1541.

35.

George

, Ruiz-Elizalde

, Beck

: A continuous subcutaneous insulin infusion needle break. Clin Diabetes, 2015; 33:195–197.

36.

Moser

, Maurer

, Binder

, Meraner

, Steichen

, Abt

, Freund-Unsinn

, Hofer

: Needle detachment in a slim and physically active child with insulin pump treatment. Pediatr Diabetes, 2015 July 23 [Epub ahead of print]. doi: 10.1111/pedi.12297.

37.

Plager

, Murati

, Moran

, Sunni

: Two case reports of retained steel insulin pump infusion set needles. Pediatr Diabetes, 2015; 17:160–163.

38.

Binder

, Lange

, Edlinger

, Meraner

, Abt

, Moser

, Steichen

, Hofer

: Frequency of dermatological side effects of continuous subcutaneous insulin infusion in children and adolescents with type 1 diabetes. Exp Clin Endocrinol Diabetes, 2015; 123:260–264.

39.

Conwell

, Pope

, Artiles

, Mohanta

, Daneman

: Dermatological complications of continuous subcutaneous insulin infusion in children and adolescents. J Pediatr, 2008; 152:622–628.

40.

Schober

, Rami

: Dermatological side effects and complications of continuous subcutaneous insulin infusion in preschool-age and school-age children. Pediatr Diabetes, 2009; 10:198–201.

41.

McVey

, Keith

, Herr

, Sutter

, Pettis

: Evaluation of intradermal and subcutaneous infusion set performance under 24-hour basal and bolus conditions. J Diabetes Sci Technol, 2015; 9:1282–12891.

42.

Patte

, Pleus

, Wiegel

, Schiltges

, Jendrike

, Haug

, Freckmann

: Effect of infusion rate and indwelling time on tissue resistance pressure in small-volume subcutaneous infusion like in continuous subcutaneous insulin infusion. Diabetes Technol Ther, 2013; 15:289–294.