Unintended intramuscular insulin injection increases the variation in the insulin absorption rate, thereby increasing the risk of hypoglycemia. To reduce the risk of intramuscular injection, shorter injection needles should be tested, particularly in lean patients (13). The objective of this study was to measure the distance from skin surface to muscle fascia in lean diabetic patients and to determine whether using an experimental 4-mm needle reduces the frequency of intramuscular injections as compared with a 6-mm needle. Further, this study tested whether the backflow of simulated insulin (test medium) was unchanged using the new 4-mm needle as compared with the 6-mm needle.

We included 21 lean children (16 male) and 32 lean adults (23 male) with diabetes and a BMI Z score <0. Cutis and subcutis thickness were measured by ultrasound at three points between the umbilicus and the spina iliaca anterior superior and also at three points at the thigh, midway between the spina iliaca anterior superior and the upper edge of patella and both 2 cm above and 2 cm below this point. Two injections of 300μl sterile air were given at the abdomen and thighs with a NovoPen 3 using an experimental 4-mm and a 6-mm NovoFine needle, respectively. The needles were inserted perpendicular to the cutis without a skinfold. Tissue deposition of the air was detected by ultrasound (2). Backflow of test medium was evaluated by injecting volumes corresponding to 10 and 40 units insulin in adults.

When categorizing the distances from skin surface to muscle fascia into the categories <4 mm and >6 mm, the majority of distances were >6 mm (abdomen 67.9%, thigh 84.9%); no patients had a skin surface to muscle fascia distance at the thigh <4 mm, and no adults had a skin surface to muscle fascia distance at the abdomen <4 mm. Significantly more patients injected subcutaneously with the 4-mm needle than the 6-mm needle at the abdomen (P < 0.032) and thigh (P < 0.001). All patients except two adults and two children injected subcutaneously at the thigh with the 4-mm needle. All adults injected subcutaneously with the 4-mm needle at the abdomen. Leakage of the test medium to skin surface was negligible and independent of injection site and needle length.

We conclude that 1) 4-mm needles reduce the risk of intramuscular insulin injections without increasing the amount of backflow of insulin to the skin surface, 2) most patients can inject with the 4-mm needle without an elevated skinfold in a 90° angle at the thigh, 3) we still propose injecting in an elevated skinfold with a 45° angle when applying a 6-mm needle in lean patients (1).

This study was supported with a grant from Novo Nordisk, Copenhagen, Denmark.

Parts of this study were presented in an oral presentation at the 33rd annual meeting of the International Society for Pediatric and Adolescent Diabetes, Berlin, Germany, 26–29 September 2007.

The technical assistance of Anne Elbæk and Pia Bonde is highly appreciated. We also thank the diabetic patients who participated.

Hofman PL, Lawton SA, Peart JM, Holt JA, Jefferies CA, Robinson E, Cutfield WS: An angled insertion technique using 6-mm needles markedly reduces the risk of intramuscular injections in children and adolescents.
Diabet Med
Birkebaek NH, Johansen A, Solvig J: Cutis/subcutis thickness at insulin injection sites and localization of simulated insulin boluses in children with type 1 diabetes mellitus: need for individualization of injection technique?
Diabet Med
Tubiana-Rufi N, Belarbi N, Du Pasquier-Fediaevsky L, Polak M, Kakou B, Leridon L, Hassan M, Czernichow P: Short needles (8 mm) reduce the risk of intramuscular injections in children with type 1 diabetes.
Diabetes Care

Clinical trial reg. no. NCT00343343, clinicaltrials.gov.

C.J. and J.S. are employed at Novo Nordisk and are involved in the development of the 4-mm needles.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.