Elsevier

Diabetes & Metabolism

Volume 33, Issue 3, June 2007, Pages 220-226
Diabetes & Metabolism

Original article
One-year efficacy and safety of Web-based follow-up using cellular phone in type 1 diabetic patients under insulin pump therapy: the PumpNet study

https://doi.org/10.1016/j.diabet.2007.01.002Get rights and content

Abstract

Aim

Conventional follow-up of type 1 diabetic patients treated with continuous subcutaneous insulin infusion (CSII) was compared with intensive coaching using the Web and the cellular phone network for retrospective data transmission and short message service (SMS).

Methods

Thirty poorly controlled patients (HbA1c 7.5–10%) were enrolled in a bicenter, open-label, randomized, 12-month, two-period, crossover study. After a 1-month run-in period, 15 patients were randomly assigned to receive weekly medical support through SMS based upon weekly review of glucose values, while 15 patients continued to download self-monitored blood glucose (SMBG) values on a weekly basis without receiving SMS. After 6 months, patients crossed over to the alternate sequence for 6 additional months. Visits at the clinic were maintained every 3 months.

Results

Patients with long-standing inadequately controlled diabetes (24 ± 13 years) were included. A non-significant trend to reduction in HbA1c (–0.25 ± 0.94%, P < 0.10) and mean glucose values (–9.2 ± 25 mg/dl, P = 0.06) during the 6-month SMS sequence was observed as compared with the no-SMS period. No safety issue (hypoglycemia, glucose variability) was reported. Adherence to SMBG was not affected by the trial. Quality of life analysis suggests a significant improvement in DQOL global score, as well as the DQOL satisfaction with life subscale, during the SMS sequence.

Conclusions

Long-term telemedical follow-up of insulin pump-treated patients using a cellular phone-, SMS- and Web-based platform is feasible, safe, does not alter quality of life and associated with a trend toward improved metabolic control.

Résumé

Efficacité et sécurité à un an de la télésurveillance par Internet et téléphone GSM chez le diabétique de type 1 traité par pompe insulinique externe : étude prospective randomisée PumpNet.

Objectifs

Le suivi conventionnel de diabétiques de type 1 traités par pompe à insuline externe a été comparé à un suivi intensif utilisant le Web et le téléphone mobile permettant la transmission rétrospective des données et l'envoi de SMS (Short Message Service).

Méthodes

Trente patients insuffisamment contrôlés (HbA1c 7,5–10 %) ont été inclus dans une étude bicentrique ouverte randomisée de 12 mois, à deux périodes croisées. Après un mois d'inclusion, 15 patients ont été randomisés entre une période comportant une aide médicale hebdomadaire par SMS s'appuyant sur les données glycémiques, tandis que les 15 autres patients ont continué à télétransmettre leurs données glycémiques chaque semaine sans recevoir de SMS. Après six mois, les groupes étaient intervertis. Les consultations conventionnelles étaient maintenues chaque trimestre.

Résultats

L'étude a inclus des patients chroniquement difficiles à équilibrer (24 ± 13 ans). Une tendance non significative à une réduction de l'HbA1c (–0,25 ± 0,94 %, P < 0,10) et de la glycémie moyenne (–9,2 ± 25 mg/dl, P = 0,06) a été observée pendant la séquence de six mois sous SMS, par référence à la période sans SMS. La sécurité du système (hypoglycémies, variabilité glycémique) n'a pas été en cause. L'observance pour l'autosurveillance était inchangée pendant l'essai. Une amélioration significative du score global de qualité de vie DQOL et du sous-score mesurant la satisfaction a été observée pendant la séquence SMS.

Conclusion

Le suivi prolongé par télémédecine de patients sous pompe à insuline, à l'aide d'un téléphone mobile, d'un serveur Web et par SMS est réalisable, sûr, n'altère pas la qualité de vie et tend à améliorer le contrôle métabolique.

Introduction

The frequency of self-monitoring of blood glucose (SMBG) is a major determinant of the quality of metabolic control among patients with type 1 diabetes mellitus [1]. Moreover, the efficacy of providing adequate education and coaching was well established in the DCCT trial [1]. However, the implementation of these principles has proven difficult in real life [2]. Patients may not perform SMBG, not use information appropriately, not report values on a diary, not bring their diary. Doctors may not provide enough availability and schedules of appointments often are not relevant with real care needs. New technologies of information and communication have acquired a state of reliability and popularity that led to many recent trials. Most systems rely upon the transmission of retrospective clinical and metabolic data, whereas some experimental systems are also able to predict future glucose levels. Yet time is still to demonstrate that these technologies are both efficient and safe in helping diabetic patients to reach therapeutic targets consistently.

We conducted a randomized trial in order to assess the relevance of telecare in adult patients with type 1 diabetes under continuous subcutaneous insulin infusion (CSII). For cost-effectiveness purposes, the telecare device relied on two broadly used technologies, cellular phone for transmission of retrospective data and short message service (SMS) for immediate medical feedback. The specific aims of this study were the assessment of metabolic efficacy, safety and quality of life.

Section snippets

Research design and methods

This was a bicenter, open-label, randomized, two-period, crossover, 12-month study in which coaching patients through the GlucoNet system using SMS was compared with a conventional follow-up in two 6-month treatment periods. During both periods, the same intensive SMBG and glucose targets were required.

Subjects entered the study based on the following criteria: age  18 years, type 1 diabetes, treated with CSII with an external pump for a minimum duration of 3 months, and with insufficient

Efficacy assessments

HbA1c measurements were performed at 3-month intervals from Visit 2 through Visit 6, at either Grenoble or Toulouse University Hospital Laboratory, using a high-performance liquid chromatography method, both certified by the National Glycohemoglobin Standardization Program.

Capillary blood glucose data transmitted by the patients to the server were used to compute an average glycemic value reflecting the 1-month (V2) or 3-month period (V3–V6) preceding each visit.

Quality of life was determined

Safety assessments

Low blood glucose was arbitrarily defined as a capillary blood glucose value < 70 mg/dl. Severe hypoglycemia was defined as any episode requiring help from a third party. Frequency of hypoglycemic episodes was computed from the data transmitted by the patients, at the time intervals described above. Occurrence of severe hypoglycemia, ketosis or intercurrent illnesses was retrospectively determined by the investigator during the visits at the clinic, so as body weight, and basal and bolus insulin

Power calculations and statistical analyses

The trial was powered to detect a 0.5% difference in HbA1c between the two groups according to a crossover setting, with a power of 0.90 and an α risk of 0.05. A sample size of 31 patients was thus calculated. We intended to include 40 patients, anticipating the loss of patients during follow-up.

Per protocol analyses were performed using SPSS software (SPSS, Chicago, IL). Results are presented as means ± S.D., or as differences with 95% confidence intervals. All analyses were preceded by a

Results

We intended to enroll 40 patients, yet at the end of the 6-month inclusion period, 31 patients had been included and followed-up according to the trial design. Upon analysis, one patient did not fulfill the inclusion criteria, as the initial HbA1c value was < 7.5%, so this patient was secondarily excluded from the analysis. The following data are per protocol analyses involving 30 patients, among which, 15 had been included by the Grenoble center and 15 by the Toulouse center. Statistical

Metabolism

Overall, patients enrolled in the SMS sequence tended to improve glycemic control compared to patients without SMS communication, as measured by HbA1c and average cumulative capillary blood glucose values (Table 1), however the differences between the two treatment sequences failed to reach statistical significance. HbA1c was reduced by –0.14 ± 0.53% during the 6-month period with SMS communication, whereas it was increased by 0.12 ± 0.65% during the semester without SMS transmission (P = 0.097,

Discussion

Several telecare trials have been recently reported using various technologies in both types of diabetes, aiming at improving metabolic control through telemonitoring, or intensive telecoaching. Whether a telemedical approach of diabetes care is efficient, safe, cost-effective and realistic is still a matter of debate.

The main contributions of the PumpNet trial are as follows:

  • the GlucoNet® platform is based upon a popular technology, GSM cellular phone, that is broadly spread among diabetic

Acknowledgments

This study was supported by research grants from Direction de la Recherche Clinique, CHU Grenoble and from Agir-a-Dom. Material was provided by Roche Diagnostics France and by Palm Inc. France. We thank Florence Putod and Carine Tchokaklian for their logistical help, and Elodie Caillet for support in statistical analysis.

References (18)

  • E. Biermann et al.

    Are there time and cost savings by using telemanagement for patients on intensified insulin therapy? A randomised, controlled trial

    Comput. Methods Programs Biomed.

    (2002)
  • H.S. Kwon et al.

    Development of web-based diabetic patient management system using short message service (SMS)

    Diabetes Res. Clin. Pract.

    (2004)
  • The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus: a randomized trial

    N. Engl. J. Med.

    (1993)
  • J.F. Gautier et al.

    Are the implications of the Diabetes Control and Complications Trial (DCCT) feasible in daily clinical practice?

    Diabetes Metab.

    (1996)
  • Reliability and validity of a diabetes quality-of-life measure for the diabetes control and complications trial (DCCT)

    Diabetes Care

    (1998)
  • E. Renard et al.

    Quality of life in diabetic patient treated by insulin pumps

    Qol Newsletter

    (2002)
  • F.J. Service et al.

    Mean amplitude of glycemic excursions, a measure of diabetic instability

    Diabetes

    (1970)
  • B.P. Kovatchev et al.

    Assessment of risk for severe hypoglycemia among adults with IDDM: validation of the low blood glucose index

    Diabetes Care

    (1998)
  • H.P. Chase et al.

    Modern transmission of glucose values reduces the costs and need for clinic visits

    Diabetes Care

    (2003)
There are more references available in the full text version of this article.

Cited by (0)

View full text