Breastfeeding in women with diabetes: lower rates despite greater rewards. A population-based study

AIMS

To explore intention to breastfeed and breastfeeding rates in hospital and on discharge across women with pre-gestational or gestational diabetes mellitus, or no diabetes.

METHODS

A retrospective cohort analysis was conducted using data from four Ontario hospitals. Women who delivered a viable infant between 1 April 2008 and 31 March 2010 were included in the study. Unadjusted and adjusted odds ratios were calculated for each outcome measure and were used to compare the breastfeeding rates among women with and without diabetes.

RESULTS

After controlling for potential confounders, women with insulin-treated diabetes were less likely to intend to breastfeed, when compared with women without diabetes (adjusted odds ratio 0.49, 95% CI 0.27-0.89). In hospital, women with insulin-treated diabetes were least likely to breastfeed (odds ratio 0.42, 95% CI 0.26-0.67), followed by women with non-insulin-treated diabetes (odds ratio 0.50, 95% CI 0.26-0.96) and women with gestational diabetes (odds ratio 0.77, 95% CI 0.68-0.87) when compared with women without diabetes. On discharge, women with insulin-treated diabetes were least likely to breastfeed (odds ratio 0.38, 95% CI 0.24-0.60), followed by women with gestational diabetes (odds ratio 0.75, 95% CI 0.66-0.85); rates of breastfeeding among women with non-insulin-treated diabetes were comparable on discharge with those of women without diabetes. Women seeking care from an antenatal provider other than a physician were 2-3 times more likely to breastfeed in hospital and on discharge.

CONCLUSIONS

Women with insulin-treated diabetes had the poorest outcomes with respect to breastfeeding rates. Gestational and non-insulin-treated diabetes were associated with lower rates of breastfeeding in hospital, while gestational diabetes was additionally associated with lower breastfeeding rates on discharge.

Self-perceived health status of women three to five years after the diagnosis of gestational diabetes: a survey of cases and matched controls

OBJECTIVE

Our purpose was to determine the long-term effect on self-perceived health status and maternal perceptions of the offspring's health status after women had been labeled with a diagnosis of gestational diabetes.

STUDY DESIGN

We designed a mail survey of women who had participated in a large cohort study 3 to 5 years earlier. A tentative address list was developed for 139 women diagnosed with gestational diabetes and 406 matched controls; 33 and 89 questionnaires, respectively, were returned as "wrong address," but 106 case and 317 control addresses were potentially valid. The primary analysis compared cases and controls on the general health scale of the SF-36. Secondary outcomes included comparisons of the groups' mean results on a worry scale, ratings of the health of the child born in the index pregnancy, disability days, physician visits, diabetes preventive behaviors, and diabetes risk appraisal.

RESULTS

The usable response rates were 65 of 106 (61.3%) and 197 of 317 (62.1%), with no meaningful demographic differences found between respondents and nonrespondents. For general health, the mean score for cases was 68.9 (SD 22.34) vs 73.8 (SD 19.86) for controls, p = 0.05 (prespecified, one-tailed). After factors found to be independently related to health perception (age, race, birth place, and comorbidity) were controlled, the differences narrowed, with a mean score for cases of 70.09 versus 73.38 for controls, p = 0.11, two-tailed. Compared with controls, cases were more worried about their own health (p = 0.02, two-tailed), rated their children as less healthy (p = 0.005, two-tailed), and perceived themselves as more likely to have diabetes (p < 0.0001, two-tailed).

CONCLUSION

The diagnosis of gestational diabetes may lead to long-term changes in how women view their own health status and that of the child born during the affected pregnancy.

Association between serum 25-hydroxyvitamin D in early pregnancy and risk of gestational diabetes mellitus

AIMS

There is emerging evidence of a relationship between vitamin D insufficiency and glucose intolerance. The aim of this study was to determine whether low serum 25-hydroxyvitamin D in early pregnancy is associated with an increased risk of gestational diabetes mellitus.

METHODS

This nested case-control study examined the association between serum 25-hydroxyvitamin D and risk of gestational diabetes within a cohort of pregnant women from March 2008 to December 2009, who had undergone antenatal screening between 15 and 18 weeks gestation and subsequent glucose tolerance testing. Cases were women diagnosed with gestational diabetes and each case was matched to up to two controls without gestational diabetes on age, race and date of blood collection. Serum 25-hydroxyvitamin D was measured from stored antenatal screening samples and compared between cases and controls.

RESULTS

Of the 116 women with gestational diabetes and 219 control subjects studied, the average age was 34.3 years and 41% were of non-Caucasian race. Women with gestational diabetes had significantly lower serum 25-hydroxyvitamin D compared with control subjects (56.3 vs. 62.0 nmol/l, P = 0.018). After adjusting for gestational age and maternal weight, serum 25-hydroxyvitamin D below the top quartile (< 73.5 nmol/l) was associated with a twofold greater likelihood of gestational diabetes (adjusted odds ratio 2.21, 95% confidence interval 1.19-4.13).

CONCLUSIONS

Lower vitamin D status in early pregnancy was associated with a significantly increased risk of subsequent gestational diabetes that was independent of race, age, season and maternal weight. This study suggests that vitamin D may influence glucose tolerance during pregnancy and provides support for studies of vitamin D as a potential intervention to prevent gestational diabetes.

Gender differences in the impact of poverty on health: disparities in risk of diabetes-related amputation

AIMS

To assess the combined impact of socio-economic status and gender on the risk of diabetes-related lower extremity amputation within a universal healthcare system.

METHODS

We conducted a population-based cohort study using administrative health databases from Ontario, Canada. Adults with pre-existing or newly diagnosed diabetes (N = 606 494) were included and the incidence of lower extremity amputation was assessed for the period 1 April 2002 to 31 March 2009. Socio-economic status was based on neighbourhood-level income groups, assigned to individuals using the Canadian Census and their postal code of residence.

RESULTS

Low socio-economic status was associated with a significantly higher incidence of lower extremity amputation (27.0 vs 19.3 per 10,000 person-years in the lowest (Q1) vs the highest (Q5) socio-economic status quintile. This relationship persisted after adjusting for primary care use, region of residence and comorbidity, and was greater among men (adjusted Q1:Q5 hazard ratio 1.41, 95% CI 1.30-1.54; P < 0.0001 for all male gender-socio-economic status interactions) than women (hazard ratio 1.20, 95% CI 1.06-1.36). Overall, the incidence of lower extremity amputation was higher among men than women (hazard ratio for men vs women: 1.87, 95% CI 1.79-1.96), with the greatest disparity between men in the lowest socio-economic status category and women in the highest (hazard ratio 2.39, 95% CI 2.06-2.77 and hazard ratio 2.30, 95% CI 1.97-2.68, for major and minor amputation, respectively).

CONCLUSIONS

Despite universal access to hospital and physician care, we found marked socio-economic status and gender disparities in the risk of lower extremity amputation among patients with diabetes. Men living in low-income neighbourhoods were at greatest risk.

Development of tertiary hyperparathyroidism after phosphate supplementation in oncogenic osteomalacia

Oncogenic osteomalacia is a rare paraneoplastic syndrome. It is characterized by bone pain, muscle weakness, gait disturbance, fractures and skeletal deformities. Hypophosphatemia, diminished renal phosphate reabsorption, decreased 1,25 dihydroxy Vitamin D and elevated alkaline phosphatase are the biochemical hallmarks of this disorder. Most tumors are of mesenchymal origin. We report the case of a 39-year-old woman with oncogenic osteomalacia caused by osteosarcoma of the right scapula which was unrecognized for several years. She subsequently developed tertiary hyperparathyroidism after treatment with oral phosphate and Vitamin D. This case illustrates that oncogenic osteomalacia may persist for many years before the tumor is discovered. This is because the tumors are frequently very small and are in obscure locations. The uniqueness of this case is the coexistence of hyperparathyroidism and oncogenic osteomalacia. Five other cases have been reported up to date. All patients had received phosphate supplement, ranging from 10 to 14 years prior to their diagnosis. Interestingly, our patient was on the treatment for only 2 years. The proposed mechanism is that exogenous phosphate stimulates parathyroid activity through sequestration of calcium.

Maternal glycaemic control and risk of neonatal hypoglycaemia in Type 1 diabetes pregnancy: a secondary analysis of the CONCEPTT trial

AIMS

To examine the relationship between maternal glycaemic control and risk of neonatal hypoglycaemia using conventional and continuous glucose monitoring metrics in the Continuous Glucose Monitoring in Type 1 Diabetes Pregnancy Trial (CONCEPTT) participants.

METHODS

A secondary analysis of CONCEPTT involving 225 pregnant women and their liveborn infants. Antenatal glycaemia was assessed at 12, 24 and 34 weeks gestation. Intrapartum glycaemia was assessed by continuous glucose monitoring measures 24 hours prior to delivery. The primary outcome was neonatal hypoglycaemia defined as glucose concentration < 2.6 mmol/l and requiring intravenous dextrose.

RESULTS

Neonatal hypoglycaemia occurred in 57/225 (25.3%) infants, 21 (15%) term and 36 (40%) preterm neonates. During the second and third trimesters, mothers of infants with neonatal hypoglycaemia had higher HbA_1c [48 ± 7 (6.6 ± 0.6) vs. 45 ± 7 (6.2 ± 0.6); P = 0.0009 and 50 ± 7 (6.7 ± 0.6) vs. 46 ± 7 (6.3 ± 0.6); P = 0.0001] and lower continuous glucose monitoring time-in-range (46% vs. 53%; P = 0.004 and 60% vs. 66%; P = 0.03). Neonates with hypoglycaemia had higher cord blood C-peptide concentrations [1416 (834, 2757) vs. 662 (417, 1086) pmol/l; P < 0.00001], birthweight > 97.7th centile (63% vs. 34%; P < 0.0001) and skinfold thickness (P ≤ 0.02). Intrapartum continuous glucose monitoring was available for 33 participants, with no differences between mothers of neonates with and without hypoglycaemia.

CONCLUSIONS

Modest increments in continuous glucose monitoring time-in-target (5-7% increase) during the second and third trimesters are associated with reduced risk for neonatal hypoglycaemia. While more intrapartum continuous glucose monitoring data are needed, the higher birthweight and skinfold measures associated with neonatal hypoglycaemia suggest that risk is related to fetal hyperinsulinemia preceding the immediate intrapartum period.

Modelling potential cost savings from use of real-time continuous glucose monitoring in pregnant women with Type 1 diabetes

AIM

To investigate potential cost savings associated with the use of real-time continuous glucose monitoring (RT-CGM) throughout pregnancy in women with Type 1 diabetes.

METHODS

A budget impact model was developed to estimate, from the perspective of National Health Service England, the total costs of managing pregnancy and delivery in women with Type 1 diabetes using self-monitoring of blood glucose (SMBG) with and without RT-CGM. It was assumed that the entire modelled cohort (n = 1441) would use RT-CGM from 10 to 38 weeks' gestation (7 months). Data on pregnancy and neonatal complication rates and related costs were derived from published literature, national tariffs, and device manufacturers.

RESULTS

The cost of glucose monitoring was £588 with SMBG alone and £1820 with RT-CGM. The total annual costs of managing pregnancy and delivery in women with Type 1 diabetes were £23 725 648 with SMBG alone, and £14 165 187 with SMBG and RT-CGM; indicating potential cost savings of approximately £9 560 461 from using RT-CGM. The principal drivers of cost savings were the daily cost of neonatal intensive care unit (NICU) admissions (£3743) and the shorter duration of NICU stay (mean 6.6 vs. 9.1 days respectively). Sensitivity analyses showed that RT-CGM remained cost saving, albeit to lesser extents, across a range of NICU costs and durations of hospital stay, and with varying numbers of daily SMBG measurements.

CONCLUSIONS

Routine use of RT-CGM by pregnant women with Type 1 diabetes, would result in substantial cost savings, mainly through reductions in NICU admissions and shorter duration of NICU care.

Insulin pump use compared with intravenous insulin during labour and delivery: the INSPIRED observational cohort study

AIMS

To assess the safety and efficacy of pump therapy (continuous subcutaneous insulin infusion; CSII) during labour and delivery in women with Type 1 diabetes.

METHODS

A retrospective cohort study of 161 consecutive Type 1 diabetic pregnancies delivered during 2000-2010 at Mount Sinai Hospital, Toronto, Canada. Capillary blood glucose levels during labour and delivery and time in/out of target (target: 4-6 mmol/l) were compared along with neonatal outcomes for three groups: (1) women on pumps who stayed on pumps during labour (pump/pump n = 31), (2) women on pumps who switched to intravenous (IV) insulin infusion during labour (pump/IVn = 25), and (3) women on multiple daily injections who switched to IV insulin infusion during labour (MDIn = 105).

RESULTS

There were no significant differences between the mean or median glucose values during labour and delivery across all three groups, and no significant difference in time spent hypoglycaemic. However, women in the pump/pump group had significantly better glycaemic control as defined by mean glucose (5.5 vs. 6.4 mmol/l; P = 0.01), median glucose (5.4 vs. 6.3 mmol/l; P = 0.02), and more time spent in target (60.9% vs. 39.2%; P = 0.06) compared with women in the pump/IV group (after removing one outlier).

CONCLUSIONS

This study demonstrates that the continuation of CSII therapy during labour and delivery appears safe and efficacious. Moreover, women who choose to continue CSII have better glucose control during delivery than those who switch to IV insulin, suggesting that it should be standard practice to allow women the option of continuing CSII during labour and delivery.

Predictors of low diabetes risk perception in a multi-ethnic cohort of women with gestational diabetes mellitus

AIM

To determine what proportion of women with gestational diabetes underestimate their diabetes risk and identify factors associated with low diabetes risk perception.

METHODS

Participants included pregnant adult women with gestational diabetes between 2009 and 2012 across seven diabetes clinics in Ontario, Canada. Data were collected through chart review and a survey that included a diabetes risk perception question.

RESULTS

Of the 614 of 902 women (68% response rate) with gestational diabetes, 89% correctly responded that gestational diabetes increases the risk for developing diabetes. However, 47.1% of women perceived themselves to be at low risk for developing diabetes within 10 years. On multivariable analysis, BMI < 25 kg/m(2) , absent previous gestational diabetes history, absent diabetes family history and absent insulin use were appropriately associated with low diabetes risk perception. However, compared with Caucasian ethnicity, high-risk ethnicity (Aboriginal, Latin American, West Indian, South Asian, Middle Eastern, Filipino, Black, Pacific Islander) [odds ratio (OR) 2.07; 95% CI 1.30-3.31] and East and South East Asian ethnicity (OR 2.01; 1.10-3.67) were associated with low diabetes risk perception. After further adjustment for immigration, only high-risk ethnicity remained a predictor of low diabetes risk perception (OR 1.86; 1.09-3.19), whereas East and South East Asian ethnicity did not (OR 1.67; 0.86-3.22).

CONCLUSIONS

Although the majority of women recognized gestational diabetes as a risk factor for diabetes, almost half underestimated their personal high diabetes risk despite prenatal care. Furthermore, women from high-risk ethnic groups were more likely to underestimate their risk, even after adjusting for immigration. Interventions tailored to these groups are necessary to enhance perceived diabetes risk.

Short-term risk of cancer among women with previous gestational diabetes: a population-based study

AIMS

To evaluate the relationship between gestational diabetes (GDM) and incidence of cancer in women within the first decade postpartum.

METHODS

This population-based retrospective cohort study compared the risk of cancer in women with GDM with that of a matched control group comprising pregnant women without diabetes. We included women from Ontario, Canada aged 20-50 years with no history of cancer who had given birth between 1995 and 2008 (N = 149 049). Women with GDM (N = 49 684) were matched on age and year of giving birth, in a ratio of 1:2, to pregnant women without diabetes (N = 99 365).

RESULTS

Over a median 8-year follow-up, there were a total of 2927 (1.5%) cancers. After adjustment for covariates, we found no significant difference in overall risk of cancer between women with GDM and matched control subjects; however, GDM was associated with a significantly greater risk of thyroid cancer (adjusted hazard ratio 1.24, 95% CI 1.05, 1.46) and a significantly lower risk of premenopausal breast cancer (hazard ratio 0.86, 95% CI 0.75, 0.98) compared with matched control subjects.

CONCLUSIONS

This large population-based study did not find a greater risk of cancers among women with GDM during the first decade postpartum; however, GDM was associated with a higher risk of thyroid cancer and a lower risk of premenopausal breast cancer. Further studies are needed to confirm these findings.

Familial hyperparathyroidism in association with colonic carcinoma

A unique case of familial hyperparathyroidism associated with carcinoma of the colon is presented. Two brothers presented initially with colonic carcinoma and years later both were found to have primary hyperparathyroidism on the basis of parathyroid hyperplasia. This raises the issue of associated malignancies in patients with hyperparathyroidism, especially if they are found to be familial. One member of the family developed severe, recurrent hypercalcemia with bone disease, and thus the need for continued follow-up is emphasized.

Managing type I diabetes in pregnancy: how near normal is necessary?

OBJECTIVE

This was a feasibility study to determine if pregnant women with type I diabetes managed with liberal target glucose values will have a decreased frequency of hypoglycemia with no differences in adverse outcomes compared with tightly controlled subjects.

STUDY DESIGN

Twenty-two women who had type I diabetes were randomized in first trimester to 'rigid' and 'less rigid' groups. Participants recorded blood glucose results and symptoms of hypoglycemia on memory-based meters.

RESULTS

Mean maternal glucose was significantly greater in first and second trimesters among patients in the 'less rigid' group. Both subjective and objective hypoglycemias were more frequent in the 'rigid' group. There were no differences between groups in cesarean deliveries, birth weights and neonatal glucose concentrations.

CONCLUSIONS

Utilizing glucose targets higher than those conventionally recommended in pregnancies of women who have type I diabetes may decrease maternal hypoglycemia while not increasing maternal or perinatal morbidity. The findings of this study justify further investigation with a larger patient base.

A composite silent corticotroph pituitary adenoma with interspersed adrenocortical cells: case report

OBJECTIVE AND IMPORTANCE

A case report of an extraordinary sellar pituitary tumor composed of corticotrophs and adrenocortical cells is presented. To our knowledge, this is only the second one reported in the literature.

CLINICAL PRESENTATION

An 18-year-old female patient presented with amenorrhea.

INTERVENTION

Investigations revealed a sellar mass, which was excised transsphenoidally. Histologically, two cell types could be readily distinguished, i.e., small basophilic cells that were positive for periodic acid Schiff and adrenocorticotropic hormone and large cells with abundant, slightly vacuolated, eosinophilic cytoplasm that were negative for periodic acid Schiff and adrenocorticotropic hormone. The nature of the tumor was revealed by ultrastructural examination, thus highlighting the importance of this technique in the investigation of pituitary adenomas. Immunohistochemistry with a panel of steroidogenic dehydrogenases and hydroxylases was positive in the large cells, confirming these as adrenocortical cells.

CONCLUSION

We suggest that the designation "composite silent corticotroph pituitary adenoma with adrenocortical cells" is an appropriate name for this tumor. The explanation for the presence of the two cell types is obscure. Two theories are considered, as were proposed by the authors of the previous case report regarding the same entity, i.e., 1) the possibility of misplaced embryonic adrenocortical cells and 2) the presence of uncommitted stem cells that differentiate into adrenocortical cells.

Continuous glucose monitoring in pregnant women with Type 1 diabetes: benefits for mothers, using pumps or pens, and their babies

AIMS

To review the current literature on the use of continuous glucose monitoring during pregnancy in women with Type 1 diabetes.

METHODS

We searched the literature for randomized controlled trials using continuous glucose monitoring during pregnancy in women with Type 1 diabetes.

RESULTS

Three randomized trials were found and discussed in this review. One UK study found a reduction in large-for-gestational-age infants; however, only masked continuous glucose monitoring was used in that study. A Danish study used intermittent real-time continuous glucose monitoring and found no differences. The present authors conducted the CONCEPTT trial, in which pregnant women and women planning pregnancy were randomized to receive continuous glucose monitoring or standard care. We found a greater drop in HbA_1c , more time spent in the target range, and a reduction in some adverse neonatal outcomes in women using continuous glucose monitoring. Numbers-needed-to-treat to prevent a large-for-gestational-age infant, a neonatal intensive care unit admission for >24 h, and a neonatal hypoglycaemia event were low. These findings were seen in both injection and pump users and across all countries. Possible reasons for differences in study findings are discussed. In addition, several issues need further study. Glycaemic variability and differences in dietary intake may also have played a role. Despite excellent glycaemic control, babies continue to be large. More research is needed to understand the role of glucose targets and the dynamic placental processes involved in fetal growth.

CONCLUSIONS

The use of continuous glucose monitoring in women with Type 1 diabetes in pregnancy is associated with improved glycaemic control and neonatal outcomes. Further research examining the glycaemic and non-glycaemic variables involved in fetal growth and the cost-benefit of using continuous glucose monitoring in pregnancy is warranted.

Dietary intakes of women with Type 1 diabetes before and during pregnancy: a pre-specified secondary subgroup analysis among CONCEPTT participants

AIM

To describe the dietary intakes of women with Type 1 diabetes before and during pregnancy.

METHODS

This was a pre-specified subgroup analysis of CONCEPTT involving 63 women planning pregnancy and 93 pregnant women from 14 sites in England, Scotland and Ireland. Two hundred and forty-six 3-day food diaries (104 planning pregnancy, 142 pregnant) were matched to data source and food reference codes, and analysed using dietary software. Participants were informed that food diaries would be de-identified and used only for research purposes.

RESULTS

Mean (sd) daily energy intake was 1588 (346) kcal and 1673 (384) kcal in women planning pregnancy and pregnant women respectively. Total carbohydrate intake was consistent with dietary guideline recommendations [180 (52) g planning pregnancy, 198 (54) g pregnant], but non-recommended sources (e.g. sugars, preserves, confectionery, biscuits, cakes) contributed to 46% of total daily carbohydrate intake. Fat consumption exceeded guideline recommendations [70 (21) g planning pregnancy, 72 (21) g pregnant]. Fibre [15.5 (5.3) g planning pregnancy, 15.4 (5.1) g pregnant], fruit and vegetable intakes [3.5 (2.2) and 3.1 (1.8) serves/day] were inadequate. Twelve women planning pregnancy (19%) and 24 pregnant women (26%) did not meet micronutrient requirements.

CONCLUSIONS

The diets of pregnant women from England, Scotland and Ireland are characterized by high fat, low fibre and poor-quality carbohydrate intakes. Fruit and vegetable consumption is inadequate, with one in four women at risk of micronutrient deficiencies. Further research is needed to optimize maternal nutrition for glycaemic control and for maternal and offspring health.

Factitious hyperinsulinemic hypoglycemia: confirmation of the diagnosis by a species-specific insulin radioimmunoassay

Surreptitious self-administration of insulin is an important cause of hypoglycemia. A 28-year-old female hospital ward clerk presented with hypoglycemia associated with an elevated plasma insulin level and a low plasma C-peptide concentration. Factitious illness was denied by the patient until it was definitively proven by using a species-specific insulin radioimmunoassay that the type of insulin circulating at the time of hypoglycemia was of animal rather than of human origin. The differential diagnosis of hypoglycemia associated with hyperinsulinemia and the current laboratory methods which may be employed to distinguish between factitious hypoglycemia and endogenous hyperinsulinism are discussed.

Effect of multiparity and ethnicity on the risk of development of diabetes: a large population-based cohort study

AIMS

To investigate the relationship between increasing parity and diabetes in a large, population-based cohort, and to examine if this relationship is different among high-risk ethnic groups.

METHODS

A population-based, retrospective cohort study was performed in 738 440 women aged 18-50 years, who delivered babies in Ontario between 1 April 2002 and 31 March 2011. Diabetes incidence postpartum was calculated for each parity and ethnic group. A multivariable analysis of the effect of parity and ethnicity on the incidence of diabetes was performed using a Cox proportional hazards model, adjusting for confounders.

RESULTS

The diabetes incidence rate per 1000 person-years was 3.69 in women with 1 delivery, 4.12 in women with 3 deliveries and 7.62 in women with ≥5 deliveries. Women with ≥3 deliveries had a higher risk of developing diabetes compared with women with 1 delivery [adjusted hazard ratios 1.06 (95% CI 1.01-1.11) for 3 deliveries, 1.33 (95% CI 1.25-1.43) for 4 deliveries and 1.53 (95% CI 1.41-1.66) for ≥5 deliveries). A similar rise in risk could be seen in Chinese and South-Asian women, with the most influence in Chinese women [hazard ratio 4.59 (95% CI 2.36-8.92) for ≥5 deliveries].

CONCLUSIONS

There was a positive and graded relationship between increasing parity and risk of development of diabetes. The influence of parity was seen in all ethnicities. This association may be partly related to increasing weight gain and retention with increasing parity, or deterioration in β-cell function. This merits further exploration.

Undiagnosed type 2 diabetes during pregnancy is associated with increased perinatal mortality: a large population-based cohort study in Ontario, Canada

AIM

To compare perinatal outcomes in women with undiagnosed diabetes with gestational diabetes alone, pre-existing diabetes and women without diabetes, and to identify risk factors which distinguish them from women with gestational diabetes alone.

METHODS

This population-based cohort study included administrative data on all women who gave birth in Ontario, Canada, during 2002-2015. Maternal/neonatal outcomes were compared across groups using logistic regression, adjusting for confounders. A nested case control study compared women with undiagnosed type 2 diabetes with women with gestational diabetes alone to determine risk factors that would help identify these women.

RESULTS

Among 995 990 women, 68 163 had gestational diabetes (6.8%) and, of those women with gestational diabetes,1772 had undiagnosed type 2 diabetes (2.6%). Those with undiagnosed type 2 diabetes were more likely to be older, from a lower income area, have parity > 3 and BMI ≥ 30 kg/m² compared with gestational diabetes alone. Infants had a higher risk of perinatal mortality (OR 2.3 [1.6-3.4]), preterm birth (OR 2.6 [2.3-2.9]), congenital anomalies (OR 2.1 [1.7-2.5]), neonatal intensive care unit admission (OR 3.1 [2.8-3.5]) and neonatal hypoglycaemia (OR 406.0 [357-461]), which were similar to women with pre-existing diabetes. The strongest predictive risk factors included early gestational diabetes diagnosis, previous gestational diabetes and chronic hypertension.

CONCLUSIONS

Women diagnosed with gestational diabetes who develop diabetes within 1 year postpartum are at higher risk of adverse pregnancy outcomes, including perinatal mortality. This highlights the need for earlier diagnosis, preferably pre-pregnancy, and more aggressive treatment and surveillance of suspected type 2 diabetes during pregnancy.