Current Screening and Management Practices for Gestational Diabetes Mellitus in Public Hospitals of South India

Introduction

Gestational Diabetes Mellitus (GDM) is recognized as any degree of glucose intolerance appearing for the first time during pregnancy.¹ As per the International Diabetes Federation (IDF), more than five million women are diagnosed with GDM annually in India.² Mothers with GDM are at increased risk of developing complications during pregnancy, including pregnancy-induced hypertension, polyhydramnios, and premature rupture of membrane, instrumental delivery, and postpartum haemorrhage. Their neonates have a higher risk of increased birth weight and hypoglycaemia.^3,4 Additionally, there is an increased risk of adverse outcomes, including developing Type 2 Diabetes Mellitus (T2DM) in later years of life.^5-7

Several international GDM diagnostic criteria are available for use in India, namely, World Health Organization (WHO), International Association of Diabetes and Pregnancy Study Groups (IADPSG) and American Diabetes Association (ADA) among others.⁸ Using a 75-gram Oral Glucose Tolerance Test (OGTT), GDM diagnosis is made when the cut-off fasting blood sugar is ≥92 mg/dL, or 2-hour postprandial blood sugar is ≥153 mg/dL. The ADA classification suggests an additional one-hour cut-off value of 180 mg/dL.^9,10 The Government of India recommends using the Glucose Challenge Test (GCT) for GDM screening, irrespective of the fasting state. This single-step 75-gram oral glucose test aids in GDM diagnosis when the blood sugar level is ≥140 mg/dL.¹¹ Similarly, in the two-step approach, initially pregnant women undergo 50-gram oral glucose challenge test (GCT), which is irrespective of the fasting status. If plasma glucose value is ≥130 mg/ dL, the woman will undergo 100-gram OGTT. The cut-off values for fasting, 1 hour, 2 hour and 3 hour blood glucose are ≥95 mg/dL, ≥180 mg/dL, ≥155 mg/dL and ≥140 mg/dL, respectively. Diagnosis of GDM follows if at least two values meet or exceed the plasma glucose concentration.¹²

In a previous study, we reported that doctors in public hospitals in India are not usually aware of these guidelines.¹³ In addition to it, confusion around diagnostic criteria, inadequate community awareness, and difficulty in access are the common reasons for the varying prevalence of GDM (8 to 41.9%).^14,15 There is scarce evidence regarding the extent to which women attending public hospitals are screened and treated for GDM.¹⁶ Specifically, we do not know what proportions of women undergo screening, are diagnosed, and receive treatment. Thus, we aimed to understand the current GDM screening and management practice in public hospitals through the study of pregnant women in Bengaluru, South India.

Materials and Methods

Study setting, participants, and data collection

We conducted a cross-sectional study in six major public hospitals where large number of pregnant women and postnatal mothers visit for health care services. Pregnant women and postnatal mothers (within seven days of their delivery) were the study subjects for cross-sectional study conducted during September 2017. We recruited all consenting pregnant women at the Ante Natal Check-up (ANC) outpatient department, irrespective of their gestational age (ranged from 5.4 - 39.1 weeks), also the mothers who had delivered in the selected public hospitals, and mothers who were available in the hospital, aged above 18 years. Further, we examined results from a contemporary on-going cohort study to understand current GDM management practices. This cohort titled, “Maternal antecedents of adiposity and studying the transgenerational role of hyperglycemia and Insulin (MAASTHI),” was set up at public hospitals in Bengaluru, South India. The details regarding the study, including the protocol for recruitment and methodology, has been published elsewhere.¹⁷ In the MAASTHI cohort study, we recruited pregnant women after completion of their 14 weeks of gestational age with obtained written informed consent. We collected baseline information and scheduled OGTT to screen for GDM upon completion of their 24 weeks of gestation. We followed up mother and children, soon after the delivery, at 14 weeks of the baby and annually until four years of the baby, to assess for adiposity of the children and several other parameters from mother-children pairs during each follow up. Characteristics of the study subjects in cross sectional study and MAASTHI cohort were similar in terms of area of residence (urban area), age group, level of education, employment and household income. Even though the cross-sectional study had postnatal mothers as study samples, only their pregnancy details were collected.

Measurements

In the cross-sectional study conducted in six public hospitals, we used questionnaires to assess the existing screening practices of GDM. We approached pregnant women and postpartum mothers in these hospitals and completed a baseline assessment after obtaining written informed consent. Information on age, education, Last Menstrual Period (LMP), parity, number of visits to ANC, screening test/s were collected. We also elicited information on knowledge regarding normal glucose levels. We assessed current screening practices in these public hospitals by identifying the glucose test prescribed to the participant and test undergone by the participant and cross-checked in the medical report. Further, we elicited information regarding, if they underwent the test in fasting state or not, whether a glucose drink was given during the procedure, the mode of blood collection for the test, number of blood samples drawn and the time interval. We compared the results of GDM screening in five of these hospitals with a tertiary public hospital wherein a cohort of women were studied during the same time.

In the MAASTHI cohort study, we recruited pregnant women and scheduled OGTT for all participants upon completion of 24 weeks of gestation. The GDM diagnosis was made using the criteria suggested by the World Health Organization (WHO).¹⁰ We assessed fasting and two-hour postprandial glucose levels after ingestion of 75 grams of glucose. The estimation of the plasma glucose levels was done in a central and accredited laboratory. Quality control samples were regularly sent to the laboratory, and all test reports were documented and cross-checked by senior team members with the reference values provided in the quality control samples. We collected self-reported data regarding the treatment received for GDM, and follow-up glucose tests underwent (if any), and this information was verified using medical reports. We obtained additional information regarding insulin therapy, other pharmacological management, and dietary advice received from the healthcare providers. We also measured the Capillary Blood Glucose (CBG) levels of mothers diagnosed with GDM, during delivery and 14-week follow-up period using the finger-prick method with the ONE TOUCH Select Simple™ Blood glucose monitoring system (022-235 Rev. Date: 04/2016 ©2010-2016 Life Scan, Inc.).

In order to assess the GDM screening cascade, we identified among the pregnant women visiting a public facility for ANC, the number of women undergoing any screening test for GDM, the number of women who underwent GCT/OGTT, and the total number of women diagnosed with GDM. The GCT is done in non-fasting state, i.e. when a pregnant woman visits health facility for ANC, she will be given 50 gram or 75-gram oral glucose solution. After two hours of consuming glucose, blood samples are collected and the blood glucose level analysed. We calculated the reduction in percentage for each of the stages of screening (any screening test, GCT/OGTT, diagnosis) until the diagnosis of GDM. Similarly, in order to investigate the GDM management cascade, we followed up women diagnosed with GDM from the MAASTHI cohort to find the number of women who received treatment, the number of these women who underwent subsequent glucose tests in order to assess treatment effectiveness (any GDM screening test undergone after receiving treatment), and finally the number of women who substantially reduced glucose levels during pregnancy and the postpartum period. We calculated the drop in percentage from the time of GDM diagnosis in pregnancy until the maintenance of optimal glucose levels in the postpartum period. We also compared the prevalence of GDM among the pregnant women who underwent OGTT in the cross-sectional study and in the on-going cohort (MAASTHI study).

Ethics approval

Both the on-going MAASTHI study and the cross-sectional study were reviewed and approved by the Institutional Ethical Review Board (IEC) at Bengaluru campus of Indian Institute of Public Health (Approval number IIPHHBTRCIEC-091-2015, dated 13^th November 2015 and IIPHHB-TRCIEC-120-2017, dated 24th July 2017). Only participants willing to participate voluntarily and provided written, informed consent were enrolled.

Statistical analysis

Statistical analysis was performed using SPSS version 23 (SPSS Inc., IL, and USA). Findings of descriptive analysis were presented in frequency and percentages. Cascade figures of screening and management practices of GDM were prepared using Microsoft Excel Spreadsheet Software.

Results

The baseline characteristics of women recruited to the cross-sectional study are provided in Table 1. Among the 186 participants, two-thirds (n=124/186) were in the age group of 18-25 years. Two-fifths of the mothers had completed their secondary education (n=76/186), and 8.6% of the mothers were illiterate. Almost 30% of mothers (29.6%; n=55/286) were primigravida, and

45.7% (n=85/186) were primipara and 24.7% were multiparous (n=46/186). Among the women in the cross-sectional study, 131/186 (70%) had undergone at least one type of glucose screening test. However, only 39/186 (22%) had undergone either OGTT or GCT. Overall, 8/186 (4.3%) of women in the cross-sectional study were diagnosed with GDM (Figure 1). Among the women who underwent GCT or OGTT, 5/39 (12.8%) were diagnosed with GDM.

In the contemporary cohort (MAASTHI cohort), we approached 1393 pregnant women to screen for GDM. Of these women, 1110 underwent a screening test for GDM and 1019 completed the test successfully. Among those who completed the test, 157/1019 (15.4%) were diagnosed with GDM. Women diagnosed with GDM in the MAASTHI cohort were followed-up to ascertain subsequent management. Among this group of women with GDM, 121/157 (77%) did not receive any GDM treatment, and 132/157 (86.6%) did not have any subsequent glucose testing in pregnancy. Only 23/157 (12.1%) had adequate control of glucose during pregnancy. In postpartum period, 19/157 (10.8%) had measured their glucose levels and only 11/157 (7%) of the GDM diagnosed women at pregnancy had adequate control of glucose during postpartum period (Figure 2).

Discussion

We have shown that a significant proportion of women with GDM are ‘missed’ from screening, diagnosis, treatment and further follow-up in public hospital settings in India. This suggests that there is poor adherence to the standard guidelines for GDM screening and management.

Our results from the MAASTHI cohort confirm the evidence suggesting a disproportionately higher burden of GDM in India compared to the pooled prevalence of 6.6% in high-income countries in Asia.¹⁸ There is substantial evidence that in low and middle-income countries, GDM is underdiagnosed and undertreated, despite the availability of low cost acceptable tests, and effective treatment. Annually there were an estimated 26 million births in India,^19,20 and around 53.5% pregnant women attend only public hospitals for antenatal care. If the true prevalence of GDM in India is around 15% as detected in our MAASTHI cohort, extrapolating this estimate suggests that among the 13 million women attending public hospitals, two million (15%) would test positive for GDM. However, in our cross-sectional study, we report that only four percent of the women attending ANC in public hospitals are diagnosed with GDM, suggesting that an additional 11% may have GDM, but remain undiagnosed due to suboptimal screening practices. This is equivalent to a staggering 1.4 million women with undiagnosed GDM receiving ANC in public hospitals in India each year. Furthermore, our results from the MAASTHI cohort suggest that among those who are diagnosed with GDM in the public hospital system, inadequate treatment and poor followup is likely to predispose them to adverse consequences. Most women with GDM can be treated in the primary care setting, and better quality of life can ensue in women and their infants if treatment were to be sustained.²¹

Several agencies recommend OGTT as the gold standard test for diagnosis of GDM.^22,23 However, very few public hospitals in India use this test for routine screening of GDM.¹³ The existing GDM screening guidelines in India suggest screening tests that have specific advantages for pregnant women as it can be conducted irrespective of fasting state and during any point of the gestation period, with a one-time venepuncture and fewer costs as compared to OGTT. Thus, it is expected that the obstetricians or primary care physicians would be vigilant in prescribing an accurate test for GDM screening in public hospitals. However, most doctors in public hospitals are prescribing neither GCT nor OGTT - two necessary screening tests recommended for use in the primary care setting.¹³ Among the 131/186 (70.4%) women in our cross-sectional study who had received some type of glucose testing, only 39/131 (22%) had been subjected to a GCT or OGTT. Among the subset of women in the cross-sectional study who had been screened via GCT or OGTT, the GDM prevalence was comparable to that reported in our MAASTHI cohort which incorporated universal screening using OGTT (12.8% vs. 15.4%).

A recent systematic review and meta-analysis in India found that prevalence of GDM detected through OGTT was 10.1% (range: 8.2-12.5%) compared to 7.4% (range: 5.2-10.2%) from the single-step GCT.¹⁴ Also difference in GDM prevalence in low and high income countries was found to be 9.3% and 6.6%, respectively.^18,24 Also, there is notable difference in urban and rural prevalence of GDM where urban women have a significantly (P <0.001) increased GDM prevalence compared to rural women as measured using different GDM diagnosing criteria.²⁵ The national guidelines recommend screening anytime during pregnancy, although the recommendations by IADPSG advise using OGTT at 24-28 weeks of gestation. However, our results indicate that screening using either GCT or OGTT (12.8% vs. 15.4%) provides estimates in a comparable range to using OGTT only. Hence, the universal implementation of the existing Government guidelines would likely result in detecting and treating a much higher proportion of women with GDM. Evidence suggests that more than half of doctors in India do not follow any of the recommended guidelines regarding GDM management,^14,26 with only six percent of the women diagnosed with GDM being promptly followed-up.²⁷ As a result, the cascade of screening and management is interrupted multiple times, increasing the risk of adverse maternal and perinatal outcomes.^28-31

The strengths of our study include the ability to assess the GDM screening methods in multiple public hospitals using comprehensive assessment of screening methods, and the opportunity to compare the ascertainment of GDM under ‘routine’ care with the gold standard for GDM diagnosis in a comparable cohort study. The limitations include the observational nature of the study with limited sample size for the cross-sectional study, and the potential influence of unmeasured confounders. Also, only a small number of pregnant women underwent OGTT/GCT in the cross-sectional study, limiting the comparison with the MAASTHI cohort where use of OGTT/GCT was widespread. We included both pregnant women and postnatal mothers in the cross-sectional study and collected pregnancy details from postnatal mothers. This may have led to differences in the accuracy of reporting due to recall bias. The comparison for GDM prevalence was obtained from an on-going cohort in a public hospital. Even though both the studies recruited study samples at various gestational ages, it would not have influenced our results since we focused only on collecting information regarding their screening tests for GDM. Although it is comparable across the public hospitals in major cities, it may not be generalized to the prevalence in rural areas. Even though it limits comparison, our estimate is within the pooled prevalence for GDM in India varying from 8-18%. This comparison is done to provide the nearest counterfactual ideal contrast, in that, what would be the prevalence of GDM in the same population if were to follow standard guidelines in the same time frame.

Conclusions

This study provides a unique insight into the current screening and management practices for GDM in public hospitals in South India. Our results suggest that most women with GDM attending public hospitals may be “missed” from GDM diagnosis and treatment, even in the major public hospitals in Bengaluru. Also, as little as one-third of women with GDM are identified by current routine screening practices in public hospitals. In addition to this considerable burden of under diagnosis, women who are diagnosed with GDM receive sub-optimal follow-up care. Thus, it is essential to reorient the public health sector regarding the importance of early screening and management to prevent early and future adverse health outcomes. Prioritizing and adhering to the standard guidelines while providing routine antenatal care services in urban areas is very much essential.

Disclosure

The authors declare that they have no competing interests.

Funding

This work was supported by the Wellcome Trust DBT India Alliance Senior Fellowship [Grant No. IA/ CPHS/20/1/505278] awarded to Giridhara R. Babu. The

funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data, preparation or approval of the manuscript; or decision to submit the manuscript for publication

Acknowledgments

We sincerely thank Department of Health and Family Welfare, Government of Karnataka (DHFW, GoK) and the Bruhat Bengaluru Mahanagara Palike (BBMP) for permitting us to conduct the study and for constant support. Our sincere thanks to Dr Giridhara R Babu for the overall support, guidance, securing grant, and management of the project and Welcome Trust, DBT India Alliance for the support with funding and guidance. We thank hospitals under DHFW, GoK, Superintendents, Medical Officers, doctors and all the support staffs of the department for their support in ongoing study. Our sincere thanks to Dr. Suresh Shapeti and Mr. T.S. Ramesh for facilitating the administrative support and coordination. We thank research team members Mr. Gurulingaiah, Ms. Maithili Karthik, Ms. Prafulla Shriyan, Mr. Kiran Kumar HN, Ms. Keerthi Deshpande, Ms. Sindhu Gowda, Ms. Meena, for their support in carrying out research activities in the field. We thank all participants for their efforts to enroll and continuous participation in the study.