Nurse-Based Management in Patients With Gestational Diabetes

On 1 July 1997, the metabolic management of women with GDM was switched from endocrinologist-based to diabetes nurse-based care without any other modification in the previously described management protocol (metabolic goals and methods, obstetric and neonatal management) (12) (Table 1). The diabetes nurse’s role in the present study was similar to that of an advanced practice nurse in the U.S. The diabetes nurses providing care to women with GDM in our center were highly trained registered nurses, with long-term specialization in diabetes (meeting qualifications for the accredited program of the Federation of European Nurses on Diabetes, as well as being employed full-time as diabetes nurses, with their tasks including structured individual and group diabetes education, diet and insulin therapy modification in patients with type 1 and type 2 diabetes, and care of patients with metabolic emergencies). Their participation in diabetes and pregnancy management began as an educational role in the care of women with prepregnancy and gestational diabetes (Table 1; “Endocrinologist-based management”) as well as participating with the obstetrician in the weekly discussion sessions. When a modification in the management protocol of women with GDM was decided in 1997, diabetes nurses were already well familiar with the metabolic goals and instruments of GDM management. The nurses visited the patients in a room adjacent to where the endocrinologist was attending women with thyroid disorders and/or prepregnancy diabetes before, during, and after pregnancy. This allowed the nurses to discuss issues related to management whenever required. Specifically, nurses consulted the endocrinologist whenever diet and/or insulin modification was not straightforward, always before initiating insulin therapy, and whenever any medically relevant issue arose.

In 2000, we undertook an evaluation of the equivalence of the two care programs. We retrospectively analyzed the equivalent-length periods of diabetes nurse-based management and endocrinologist-based management using the database that is prospectively and continuously collected. Chart review, data entry, and analysis were performed regularly by the endocrinologists shortly after delivery as part of a continuous quality improvement strategy. Three forms (one each for pregnancy, delivery, and neonatal outcome) were used for data collection and later entered in an Access (Microsoft) database. Internal validation rules and double checking of each entered record were used to ensure data accuracy.

The diagnosis of GDM was established by universal screening and oral glucose tolerance test (OGTT) in accordance with the first three Workshop Conferences on Gestational Diabetes Mellitus (13), as described in reference 12. We previously reported a GDM prevalence of 12% with this screening protocol (14).

All women diagnosed with GDM attending the Diabetes and Pregnancy Clinic from 1 January 1995 to 31 December 1999 were included in the study. Until 30 June 1997, women with GDM received endocrinologist-based metabolic management (n = 244: 232 singleton, 11 twin, 1 triplet), whereas from 1 July 1997 to 31 December 1999, they received diabetes nurse-based metabolic management (n = 283: 272 singleton, 9 twin, 2 triplets).

Therapy was initiated with normocaloric diet and self-monitoring of blood glucose, with insulin therapy being added when necessary (12) to achieve glycemic target values (90 mg/dl fasting/preprandial and 120 mg/dl postprandial) (15,16).

Hypertension was defined as blood pressure ≥140/90 mm/Hg, recorded on two occasions at least 6 h apart (17), as well as all variables of perinatal outcome previously described (12).

Baseline characteristics, insulin treatment, and perinatal outcome of women with GDM in the two periods were compared. A χ² test, with Fisher’s correction, was used to analyze categorical data. As quantitative data did not adjust to a normal distribution (Kolmogorov-Smirnov test), they were expressed as median (range); the Mann-Whitney U test was used for analysis. Significance was set at P < 0.05, with Bonferroni correction for 9 variables in the case of maternal characteristics (P < 0.006) and 15 variables in the case of outcome (P < 0.0033).

Maternal characteristics included the following (percent or median [range]): age, 33 years (19–44); BMI, 22.95 kg/m² (15.4–43.9); family history of diabetes, 57.7%; prior history of hyperglycemia, 17.2%; gestational age at diagnosis, 32 weeks (3–40); fasting blood glucose in diagnostic OGTT, 4.5 mmol/l (3.3–10.9); 1-h blood glucose in diagnostic OGTT, 11.5 mmol/l (5.5–18.7); 2-h blood glucose in diagnostic OGTT, 10.2 mmol/l (4.8–18.7); and 3-h blood glucose in diagnostic OGTT, 7.5 mmol/l (1.8–23.7). Characteristics did not differ for mothers in the two study periods (Table 2).

The rate of insulin treatment was 38.2%; maternal hypertension, 6.4%; preterm delivery, 10.0%; cesarean section, 29.4%; 1-min Apgar <7, 5.0%; macrosomia, 4.0%; small-for-gestational-age newborns, 8.3%; large-for-gestational-age newborns, 7.0%; obstetric trauma, 2.4%; major malformations, 4.5%; hypoglycemia, 1.5%; hypocalcemia, 0.5%; polycythemia, 2.8%; jaundice, 4.5%; respiratory distress, 3.2%; and mortality, 1.5%. No differences were seen between mothers treated in the two periods (Table 3).

Because of the nonsignificant increase in mortality in the period of diabetes nurse-based management (2.0 vs. 0.8%; NS), the causes of mortality in both periods were analyzed. The two cases of neonatal mortality in the period of endocrinologist-based management were attributed to hyaline membrane in a preterm twin (1) and a Fallot tetralogy (1); the six cases in the period of diabetes nurse-based management were attributed to stillbirth in a twin fetus (2), preterm birth before 24 weeks (2), respiratory distress secondary to meconium aspiration (1), and cardiac malformation in a baby with chromosome 13 trisomy (1).

Most guidelines and original studies on the care of women with GDM do not describe the role of the diabetes nurse as an active one, whereas it is clearly stated that diet instruction should be carried out by dietitians (6,9). Nurse practitioners today are increasingly furthering their training, developing specialized skills, and making autonomous diagnoses and treatment decisions, with outcomes that do not differ from those of physicians (18,19). Their wider role is becoming increasingly recognized not only in diabetes management but in many fields (20,21), such as in emergency units (22), primary care (18,19,23), and pediatrics (24), and has been the subject of several recent systematic reviews (19,21). However, to the best of our knowledge, there has been no study testing the performance of nurse practitioners or advanced practice nurses in the care of women with GDM.

The present study was not a controlled clinical trial, as it was conceived retrospectively. However, as data had been collected prospectively before the research question arose, the study should be considered nonbiased. The only change in the management protocol was the difference in the distribution of health professionals’ time: in the period of endocrinologist-based management, the endocrinologist spent more time with the woman and the diabetes nurse provided backup, whereas the reverse was true in the period of diabetes nurse-based management. The study and historical control groups displayed similar baseline characteristics and no differences either in the rate of insulin treatment or in perinatal outcome. The trend toward increased mortality in the period of nurse-based management was further analyzed, and most underlying causes (stillbirth in twin pregnancy, extreme preterm birth, and cardiac malformation in an infant with trisomy) were not related to metabolic control after GDM diagnosis. Subsequently, they cannot be attributed to diabetes nurse-based management. Treatment satisfaction was not evaluated, but we believe that endocrinologist backup in diabetes nurse-based management provides reassurance to patients, nurses, and the physicians themselves. In addition, although available for backup whenever required, the endocrinologist remained able to attend his patients.

A prospective controlled trial is necessary to give definitive proof that care provided to women with GDM by diabetes nurses and endocrinologists is equivalent. However, in the meantime, the present study supports a more active role of the diabetes nurse in the management of women with GDM.