Study design
The Diabetes Pearl is an observational cohort study, in which all eight Dutch academic medical centers participate. The initial organization of the String of Pearls Initiative in the Netherland was cofinanced by the Dutch Government and the eight Dutch academic medical centers. The continuation is financed by the academic medical centers.
Study population and inclusion criteria
Patients who were diagnosed with type 2 and who received secondary or tertiary medical care in one of the 6 academic medical centers in Amsterdam, Utrecht, Nijmegen, Rotterdam, Leiden or Groningen, primary medical care in the area of Hoorn or one of the three types of care in the region of Maastricht, were eligible for participation in the Diabetes Pearl (Figure
). Patients were excluded if their ability to understand and write the Dutch language enabled them to provide written informed consent.
Ethical considerations, patient information and written informed consent
All protocols used to create the Diabetes Pearl, and all generic String of Pearl biobanking and data-procedures were reviewed as a multi-center study by the ethics committee of the VU University Medical Center. Ethical Committies of all seven other academic centers provided positive advise for participation. To provide patients with enough time to decide whether or not they wanted to participate, study information including an informed consent form was sent by mail prior to their clinic visit. Upon arrival at the clinic, an oral explanation was provided and patients who agreed to participate, signed the informed consent form. Written informed consent included participation in the study itself as well as separate consent items covering biobanking, genetic analysis, and linkageto important national databases like the municipality register for mortality follow-up. All participants received a copy of the signed informed consent.
Baseline data collection of the Diabetes Pearl
Since the start of the String of Pearls Initiative in 2007, all recruitment centers have worked together to create the optimal dataset needed to gather baseline data of patients with type 2 diabetes in the Netherlands. This minimal dataset was supported by local coordinators of the participating clinics and introduced accordingly. In addition, work instructions were translated into standard operational procedures (SOPs) to ensure comparability of all data collected.
Baseline examinations began November 2009 and will continue through 2012. From all patients who provide informed consent, the following information was collected:
1. Demographics and medication use
Hospital information systems at all recruitment centers were used to collect information on date of birth and gender. Participants were requested to bring all their medication on the day of the visit or lists from pharmacists were used so that medication use (name, indication for description, dose, frequency, ATC-code) could be reported accurately.
2. Questionnaire data
All participants received a self-report questionnaire in which the following items were included: socioeconomic status of the participants and their partner, smoking behavior, alcohol consumption habits and (family) history of diabetes, stroke and CVD. In addition, a modified version of the Rose Questionnaire for the diagnosis of ischaemic heart pain and intermittent claudication [
5] and the EuroQol 5D questionnaire for information on quality of life [
6] were included. In 7 of the 8 recruitment centers, the questionnaire was extended with the following items: information on hypoglycaemic events, the SF-12 questionnaire on quality of life [
7], the PAID Questionnaire on diabetes-related emotional distress [
8], and the DN4 questionnaire on neuropathic pain [
9].
3. Physical examinations
Weight and height were measured bare foot and wearing light clothing using a clinical stadiometer and scale. Waist circumference was measured between the lower rib margin and the spina iliaca anterior superior, and the hip circumference over the maximum of the buttocks. A tape measure was used in all recruitment centers. Two measurements of both waist and hip were performed. Final waist and hip circumference were calculated as the mean of the two measurements. Blood pressure was determined three times on the right arm after a 10
minute rest period, using a non-invasive blood pressure monitor (Omron 7051
T in seven centers, Colin Press BP 8800p in one center). Final blood pressure was calculated as the mean of the last 2 measurements.
4. Electrocardiogram
A rest 12-lead electrocardiogram (ECG) was performed on all participants. ECGs were archived electronically.
5. Fundus photography
To determine the presence of diabetic retinopathy, fundus photography of both eyes (with dilated pupils in case no clear picture could be obtained) was performed in 7 of the 8 clinics. All fundus photographs were performed with a non-mydriatic camera in 45 degrees of at least two fields: one field centered on the macula and one nasal field with the optic disc positioned on a disc-diameter from the temporal edge of the field. This is in accordance with the EURODIAB protocol [
10].
6. Ankle-brachial blood pressure index
As indicator of peripheral arterial disease, we obtained an ankle-brachial blood pressure index from all participants in 7 out of the 8 centers. Blood pressures were measured at the brachial artery of the left and right arm, and twice at the dorsal pedal artery and posterior tibial artery of the left and right ankle. The ankle-brachial index was calculated for every ankle separately as the highest recorded pressure on that ankle divided by the highest pressure in the brachial artery. In 6 centers, the blood pressures were determined with a Doppler device and the index was calculated by hand. In one center, the Omron VP2000 was used, which determined the blood pressures automatically and calculated the corresponding ankle-brachial index.
7. Peripheral vibration perception
With the use of a Horwell Neurothesiometer (Scientific Laboraty Supplies, Nottingham, UK), peripheral vibration threshold was tested in 7 centers. Vibration thresholds were tested 8 times at the distal phalanx of the hallux. Mean vibration threshold was then calculated as the mean of the 6 highest vibration thresholds.
8. Laboratory measurements
Fasting venous blood plasma was used to determine fasting glucose, total cholesterol, HDL cholesterol, triglycerides and serum creatinin. A fasting whole blood sample was used for the determination of HbA1c level. In addition, a sample of morning urine was collected and used for the determination of urine albumin and urine creatinin. All the laboratories were certified and located on site in the 8 clinics.
9. Biobanking
An important part of the Diabetes Pearl is creating a national biobank. Therefore, additional samples of venous blood (serum, EDTA plasma, citrate plasma) and morning urine (including and excluding anti-oxidants) were collected for the storage of plasma, serum and urine. Samples were aliquoted and stored at -80 degrees. In addition, an EDTA whole blood sample was collected for DNA extraction. After isolation (on site in all 8 recruitement centers), a quality control on the DNA was performed and DNA was catalogued and stored in -80 degrees Celsius storage freezers.
Data processing
Since the start of the Diabetes Pearl, much attention is given to create a solid research infrastructure. As a result, the Diabetes Pearl consists of a well-designed minimal dataset and the use of standardized data collection procedures across centres. In addition, clinical data, questionnaire data, laboratory data and biobank information are stored on a national level in a highly sophisticated data system. Before the data are uploaded nationally, several data checks are performed to investigate completeness and correctness of the data. Moreover, participant’s name and address is kept only at the recruitment center, all uploaded data are pseudonomized. Biobank material is also stored anonymous and standardized storage guidelines have been developed. Biobank storage is provided local at all recruitement centers.
Preliminary results
Until February 2012, 4.587 patients were included in the Diabetes Pearl of whom 2.320 received primary medical care and 2.267 received secondary or tertiary medical care. It is to be expected that by the end of 2012, a total of 7000 patients will be included.
1,2 Miranda T Schram,3,4 Evertine J Abbink,5 Wanda M Admiraal,6 Marja W Dijk-Schaap,7 Frits Holleman,6 Giel Nijpels,1,8 Behiye Özcan,9 Hanno Pijl,7 Nicolaas C Schaper,10,11 Eric JG Sijbrands,9 Bianca Silvius,12 Cees J Tack,5 Harold W de Valk,12 Bruce HR Wolffenbuttel,13 Coen DA Stehouwer,3,4 and Jacqueline M Dekker1,2
publication history