The Professional Medical Journal

www.theprofesional.com ORIGINAL PROF-2407

GESTATIONAL DIABETES;
SENSITIVITY, SPECIFICITY AND DIAGNOSTIC ACCURACY OF
GLUCOSE CHALLENGE TEST IN DIAGNOSIS

Dr. Ayesha Snover1, Dr. Kinza Alam2, Dr. Tahir Ahmad Munir3, Dr. Rabia Sajjad4, Dr. Farhat Naz5

1. Assistant Professor
Rawal Institute of Health Sciences, ABSTRACT… Objective: To find sensitivity specificity and diagnostic accuracy of Glucose
Islamabad
2. Assistant Professor
Challenge test in diagnosing Gestational Diabetes in Pregnant women. Due to poor socio-
Rawal Institute of Health Sciences, economical and educational status, dietary habits and ignorance regarding pregnancy related
Islamabad problems probably increase the prevalence and burden of gestational diabetes mellitus (GDM)
3. Associate professor and its complications in pregnancy. Best and simple strategy to identify women with gestational
Rawal Institute of Health Sciences,
Islamabad diabetes is still lacking and unclear. Study Design: Cross sectional study. Setting and Duration:
4. Classified Gynaecologist This study was performed at Jinnah Hospital Lahore, from Nov 2005 to Dec 2006. Methodology:
CMH Sargodha A glucose challenge test (GCT) was performed on 500 selected pregnant women by giving 50-g
5. Associate professor
Jinnah Hospital Lahore
glucose in water orally. A serum glucose level ≥140 mg/dl after an hour was taken as positive test.
To confirm GDM, 75 g glucose in 200 ml of water was given and sugar levels after 2 hrs by
Glucometer, >200 mg/dl confirmed GDM. Results: An increasing trend in age, gestational age
and BMI and a significant difference regarding positive family history of diabetes and gravidity
was seen in patients with GDM compared to normal pregnant. The maximum percentage of GDM
Correspondence Address:
Dr. Kinza Alam was noted in multigravida, between 25-29 years, BMI >28kg/m2, and a gestational age of 28
House No 36A, Rose Lane 2 weeks. The sensitivity of GCT was 80%, specificity 97.8%., and diagnostic accuracy was 96.4%.
Lalazar-I, Rawalpindi Cantt Conclusions: Screening is necessary to identify women with GDM. A 50-g glucose challenge
drkinzairfan@gmail.com
test might be acceptable as a screening test for GDM as it has high sensitivity, specificity and
diagnostic accuracy.

Key words: Glucose Challenge test, Sensitivity, Specificity, gestational diabetes mellitus
Article received on:
20/12/2013
Accepted for Publication: Article Citation: Snover A, Alam K, Munir TA, Sajjad R, Naz F. Gestational diabetes; sensitivity,
03/02/2014
Received after proof reading:
specificity and diagnostic accuracy of glucose challenge test in diagnosis.
21/04/2014 Professional Med J 2014;21(2): 360-366.

INTRODUCTION and ethnicities, as well as with the diagnostic

2-5
Gestational diabetes mellitus (GDM) is defined as criteria used . Specific risk factors and the degree
any degree of glucose intolerance with onset first of their influence on GDM prevalence are difficult
recognized during pregnancy and commonly to quantify across populations. However, a
recognized after 20 weeks gestational age1. This number of clinical risk factors have been
condition is associated with increased risks for the demonstrated to be associated with an increased
fetus and newborn, including macrosomia, likelihood of GDM, including age, ethnicity, and
shoulder dystocia, birth injuries, hyperbiliru- obesity, family history of diabetes, past obstetric
binemia, and hypoglycemia, respiratory distress history and in populations with a high risk of type 2
syndrome, childhood obesity, and have an diabetes mellitus like American Indians, African
increased risk for the development of type 2 Americans, Hispanic/Latino Americans, and Asian
2,3 4-6
diabetes later in life . Maternal risks include Americans .
preeclampsia, cesarean delivery, and an
increased risk of developing type-2 diabetes However, recently it has been demonstrated in a
within 5–15 years of delivery. The prevalence number of high-quality studies that the risk of a
varies significantly among different populations number of important complications associated

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 360

GESTATIONAL DIABETES 2

with GDM, can be reduces by treatment of GDM MATERIAL AND METHODS

with diet or insulin thus improving both perinatal as This cross sectional study was carried out at
7,8
well as maternal outcome from 4% to 1% . department of obstetrics and Gynecology (unit II)
Jinnah Hospital Lahore during 2005 to 2006. The
In our country, the major portion of the population sampling technique used was convenient non
lives in rural areas with meager facilities in terms of probability technique. The pregnant women of all
health care delivery. Poor socio-economical and parity, with a gestational age of 24-28 weeks, and
educational status, dietary habits, ignorance got registered at Gynae and Obs unit II, Jinnah
regarding pregnancy and pregnancy related Hospital Lahore, were included in the study. The
problems operate unfavorably and probably pregnant women with known diabetes mellitus,
increase the burden of GDM. However social other morbid diseases like heart disease or
practices, taboos associated with pregnancy, malignancy, or taking drugs that alert blood
unauthorized practitioners and overlook of such glucose level like steroids were excluded from the
conditions even by specialists, due to lack of study. A total of 500 pregnant women were
sensitivity towards importance of this situation included in the study using convenient non
have a strong influence on the prevalence of GDM probability sampling technique.
and its complications in pregnancy.
At 1st visit we evaluate the patients for age,
Identifying women with GDM in order to provide duration of pregnancy, gravidity, any obstetric
treatment has therefore become of eminent history of all major events, relevant personal
importance, but is difficult as clinical signs and history especially dietary habits, nutritional status,
symptoms are often absent. Because of the lack of past history of any other major illness, history of
clinical signs and symptoms of GDM, screening diabetes in 1st degree relatives and complication
tests are essential to identify women with GDM. related to present pregnancy. General
One of the tests that is used in the diagnostic examination included measurement of height,
pathway is the 50-g glucose challenge test9. weight and B.P., obstetric examination, and base-
Although currently the 50-g glucose challenge test line investigations were done if necessary.
(GCT) is not recommended in the majority of
national guidelines, it could be a useful test in the Method of performing GCT
diagnostic work-up of GDM10. Since the two hour Screening test for GDM was performed by giving
oral glucose tolerance test (GTT) is a very time 50 gm. glucose orally to the patient at any time of
consuming method, but can be used as alternative the day without dietary preparations. Patient was
in patients with high risk factors. asked not to take any oral in between. The serum
glucose was measured one hour later. If the blood
Recently in line with the Expert Committee’s glucose one-hour value was =140 mg/dl the
recommendation, the World Health Organization screened test was taken as positive; and with
and the American Diabetes Association <140 mg/dl was taken as negative.
subsequently adopted a hemoglobin A1C level of
6.5 percent or higher as a new diagnostic criterion Method of performing OGTT
for diabetes. For people who do not have diabetes, This test was performed on all patients who
a normal hemoglobin A1C level is around 5%11. underwent GCT according to WHO recommenda-
tions. 75 gm of glucose was dissolved in 200 ml of
The aim of this study was to analyze the accuracy water and the patient was asked to drink it within 5
of the 50-g glucose challenge test for detection of minutes. The time was noted and the patient was
glucose intolerance in pregnancy in order to asked to come back after an hour for the test. A
evaluate its applicability as first-step screening test capillary blood specimen was obtained and tested
for GDM; with its sensitivity, specificity to current for blood sugar levels by Medisense Optimum
oral glucose tolerance test (OGTT). Glucometer that works by electrical current

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 361

GESTATIONAL DIABETES 3

produced by chemical reaction between glucose (n)and percentages (%) for nominal data.
and glucose dehydrogenase, NAD, and
phenanthelin quinine present on the glucose RESULTS
strip.If the blood sugar levels were greater than Table-I shows the demographic data of the
140mg%, the test was considered positive and patients. The mean age of the study group was
these patients were subjected to OGTT to confirm less than 25 years. Out of 500 pregnant women
the diagnosis of gestational diabetes. who underwent GCT and OGTT, only (n=40) 8
percent of pregnant women were found positive
The sensitivity, Specificity, diagnostic accuracy for gestational diabetes. Out of 40 GDM patients
and Predictive values of Glucose Challenge test only 3.4% (n=17) have positive family history of
was find out by the following formulas10. diabetes mellitus. Concerning the BMI, the GDM
patients have slightly higher but non-significant
Sensitivity Test for GCT body mass index compared to non-diabetics. The
Sensitivity Rate = 100 x True Test / (True Positive + mean gestational age was slightly higher in
False Negative) patients of GDM compared to non-GDM women.
Specificity Test for GCTs The main bulk of the study population and GDM
Specificity Rate = 100 x True Negative / (True were formed by multiparous women.
Negative + False Positive)
Diagnostic Accuracy Table-II shows demographic analysis of patients
Diagnostic Accuracy = 100 x (True Positive + True with Gestational Diabetes mellitus.
Negative) / (False Positive + False Negative)
Predictive Value for Positive Test The maximum number of patients with positive
Positive Predictive Value = 100 x True Positive / GDM has age duration between 25-29 years, while
(True Positive + False Positive) the patients above 30 years showed least
Predictive Value for Negative Test evidence of GDM. The multigravida showed a
Negative Predictive Value = 100 x True Negative / significant to develop GDM compared to
(True Negative + False Negative) pimigravida, while a less percentage (42.5%) of
GDM have a positive family history of diabetes
STATISTICAL ANALYSIS mellitus. The GDM was significantly more in
Statistical analysis was done using SPSS 17, patients having a BMI more than 28 kg/m2
Chicago, Illinois. Chi-square test was performed to compared to those with BMI <27 kg/m2. The GDM
assess the statistical significance by p values of also showed a relation with duration of gestation,
=0.05 were considered significant. The results are the patients having more gestational age have
given as mean standard deviation (SD) for more percentage of GDM.
normally distributed data and as frequencies

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 362

GESTATIONAL DIABETES 4

A total of 500 patients were screened for 97.8%., while the diagnostic accuracy was 96.4%.
gestational diabetes by glucose challenge test. The positive predictive value of GCT was 96.2%,
Only (n=40), 8% patients showed a positive GCT. while negative predictive value was 85.4%.
The oral glucose tolerance test was performed on
patients who showed positive GCT; a true positive DISCUSSION
OGTT was detected in 32 patients, while false Our result showed 8% of the patients diagnosed as
positive in 10 patients. Out of 500 patients, 458 GDM, which is in consistence with a number of
12,13
showed negative GCT, out of which 450 were true studies who reported prevalence for GDM
negative and 8 were found to be false negative by ranges from 1%-14% of all the pregnancies,
OGTT. On the whole, out of 500 patients, 40 were depending upon population studied and
diagnosed as GDM by OGTT. diagnostic criteria used. However, the results were
14
found to be contradictory with Ramirez et al who
The sensitivity of GCT was 80%, and specificity reported the GDM prevalence of 17.2%, the

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 363

GESTATIONAL DIABETES 5

difference may be due to ethnic group and tolerance during pregnancy. The most important
duration of study. of these being past history of diabetes in first
degree relatives, still birth and birth of overweight
It is universally accepted that the incidence of infant. In our study we observed that 17 out of 40
GDM is high in Asian, Mexicans, Native Americans (42.5%) cases of gestational diabetes had a
and African-American women15. positive history of diabetes in first-degree relatives.
Rhee and Catherine also reported more than 50%
Out of the 40 gestational diabetic patients, the cases of GDM with positive history of diabetes in
maximum of 20 patients (50%) belonged to the first-degree relatives respectively in their series25,26.
age group of 25-29 years and 15 patients (37.5%) However Campbell et al observed that only 9% of
were between 20 to 24 years i.e.8 7.5% of all the his GDM cases gave a history of diabetes in first-
27
gestational diabetic patients were between the degree relatives .
ages of 21 to 30 years. Coustan et al reported a
similar finding of 56% of GDM cases under 30 We in this study use GCT and then who were
years of age16. Similar conclusion has been re- positive for GCT underwent OGTT. In many
ported by Hughes et al with maternal age ranging potentially relevant studies dealing with the 50-g
from 17 years to 41 years among the positive glucose challenge test in pregnant women, the
17
cases, mean age being 29.4 years . However OGTT was only performed if the 50-g glucose
Green et al opined that there was an increased challenge test was considered to be abnormal.
incidence of GDM with increasing age18. But This design characteristic, known as partial
Granat et al reported that only 18.7% of their verification, is encountered in many studies on
patients were of older age group19. diagnostic accuracy: to minimize the burden of
possibly redundant additional testing in women
In present study it is turned out that GDM has more with a negative screening test result, only
association with larger BMI. The results are in abnormal screening test results are verified by the
agreement with Cypryk et al20 who concluded that reference test28.
women with BMI >30, have greater risk of
developing GDM. Nohira et al21 also reported that Depending on the application of the test
women with increased weight gain during (screening or alternative diagnostic) and the
pregnancy have greater chance to develop GDM consequences of false-positive and false-negative
during pregnancy. test results, certain combinations of accuracy
values are preferred. These values depend on
There is a general agreement regarding whether it is more harmful to classify women as
increasing age and parity as one of the most false-positive or false-negative, taking all possible
common risk factor for developing GDM. We consequences of such results into account. In the
observed an incidence of 35% GDM individuals in case of GDM, regarding the nature and
primigravida and 65% in multigravida. Our results consequences of the disease, one should aim for
are in agreement with a number of studies (John et an adequate detection rate, albeit not at the cost of
al Al-Rowaily et al22,23 who found a higher incidence an unacceptable false-positive rate. If the 50-g
(59.9%) of GDM in multigravida, while the results glucose challenge test is used as a screening test,
from Jawa et al24 has noticed more occurrence of a higher sensitivity rate than74% would probably
GDM in primi gravida 42.7%, however, Granat et al be warranted to accept a false-positive rate of 83%.
did not find any correlation between parity and Moreover, if one considers using the 50-g glucose
alterations of carbohydrate metabolism in their challenge test as a diagnostic test for GDM, higher
19
study . detection rates are required. As the prevalence of
GDM in the general population is relatively low, a
Various factors help in identifying women who are clinically useful test would thus have to have a high
at higher risk of developing abnormal glucose positive LR (>10)29.

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 364

GESTATIONAL DIABETES 6

A glucose loading test like the 50-g glucose glucose challenge test can be used as diagnostic
challenge test in theory seems an adequate test for GDM.
method to mimic post prandial glucose levels, and Copyright© 03 Feb, 2014.
therefore to measure the degree of glucose (in)
tolerance in pregnancy. A health technology REFERENCES
report concerning various screening strategies for 1. Hedderson MM, Gunderson EP, Ferrara A.
GDM stated that the cost-effectiveness of a Gestational Weight Gain and Risk of Gestational
Diabetes Mellitus. Am J Obstet Gynecol 2010; 115:
number of studies find that screening with the 50-g 597-604.
glucose challenge test, and then testing screen-
positives with the OGTT, was less costly than going 2. Nicholson W, Bolen S, Witkop CT, Neale D, Wilson L,
straight to universal OGTT. However, a high-quality Bass E. Benefits and risks of oral diabetes agents
compared with insulin in women with gestational
cost-effectiveness analysis developed by the UK’s diabetes: a systematic review. Obstet Gynecol
National Institute for Health and Clinical 2009;113:193–205.
Excellence (NICE) guideline development group
found that two screening strategies dominated: 3. Getahun D, Nath C, Ananth CV, Chavez MR, Smulian
selection by American Diabetes Association (ADA) JC. Gestational diabetes in the United States:
temporal trends 1989 through 2004. Am J Obstet
criteria, followed by the 75-g OGTT; and selection Gynecol 2008;198:525.e1–525.e5.
by high-risk ethnicity, followed by the 75-g OGTT.
In view of these findings and as an extension to the 4. Lawrence JM, Contreras R, Chen W, Sacks DA.
results of the cost-effectiveness analysis of the Trends in the prevalence of preexisting diabetes
and gestational diabetes mellitus among a
NICE guideline development group, it would be racially/ethnically diverse population of pregnant
interesting to consider the cost effectiveness of a women, 1999–2005. Diabetes Care
strategy that consists of selection based on 2008;31:899–904.
various risk factors, followed by screening with a
50-g glucose challenge test, followed by an OGTT 5. Metzger BE, Lowe LP, Dyer AR, Trimble ER,
Chaovarindr U, Coustan DR, et al.; HAPO Study
in the case of an abnormal test result of the 50-g Cooperative Research Group. Hyperglycaemia
glucose challenge test, and to compare this in a and adverse pregnancy outcomes. N Engl J Med
randomized controlled trial with other screening 2008; 358:1991–2002.
10
strategies .
6. Kim SY, England L, Wilson HG, Bish C, Satten GA,
Dietz P. Percentage of gestational diabetes
CONCLUSIONS mellitus attributable to overweight and obesity.
Screening is necessary to identify women with Am J Public Health 2010;100:1047–52.
GDM. High sensitivity is often warranted in
screening tests, as a false-negative test result (in 7. Crowther CA, Hiller JE, Moss JR, McPhee AJ,
Jeffries WS, Robinson JS; Australian
which disease remains undiscovered) is Carbohydrate Intolerance Study in Pregnant
considered to be more harmful than a false- Women(ACHOIS) Trial Group. Effect of treatment
positive test result (in which a reference test is of gestational diabetes mellitus on pregnancy
unnecessarily performed). A good detection rate outcomes. N Engl J Med 2005; 352:2477–86.
of the 50-g glucose challenge test might be 8. Tam WH, Ma RC, Yang X, Li AM, Ko GT, Kong AP, et al.
acceptable as a screening test for GDM. Glucose intolerance and cardiometabolic risk in
adolescents exposed to maternal gestational
Using the OGTT for screening could be a lesser diabetes: a 15-year follow-up study. Diabetes
burden and more cost-effective than a two-step Care 2010;33:1382–4.
method in which a glucose loading test might be 9. Landon MB, Spong CY, Thom E, Carpenter MW,
performed twice. As the sensitivity of GCT was Ramin SM, Casey B, et al. Eunice Kennedy Shriver
80%, specificity 97.8%., diagnostic accuracy National Institute of child health and human
96.4%; positive predictive value 96.2% and development maternal-fetal medicine units
network. A multi-center, randomized trial of
negative predictive value was 85.4% so the 50-g treatment for mild gestational diabetes. N Engl J

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 365

GESTATIONAL DIABETES 7

Med 2009;361:1339–48. type-2 diabetes in mothers. Przegl Lek

2005;62:38-41.
10. D G Altman, J M Bland. Statistics Notes:
Diagnostic tests 1: sensitivity and specificity. 21. Nohira T, Kim S, Nakai H, Okabe K, Yoneyama K.
BMJ1994;308:1552. Recurrence of gestational diabetes: rate and risk
factors from initial GDM and one abnormal GTT
11. Racial Differences in Diabetes Diagnostic value. Diabetes Res Clin Pract 2006;71:75-81.
Thresholds Could Warrant Changes in
Guidelines, Healthcare Delivery. Diabetic care 22. John K, Olynik C,Mase R, Kreisman S, Tildesley H.
2012. Gestational diabetes mellitus outcome in 394
patients. J Obstet Gynaecol Can 2006;28:122-7.
12. Zeba H, Ata-ur-Rehman M. Serum leptin level in
female patient with NIDD. J Coll Physicians Surg 23. Al-Rowaily MA, Abolfotouh MA. Predictors of
Pak 2003;3:67-9. gestational diabetes mellitus in a high-parity
community in Saudi Arabia. East Mediterr Health J.
13. Hadaegh F, Tohidi M, Harati H, Kheirandish M, 2010;16:636-41.
Rahimi S. Prevalence of gestational diabetes
mellitus in southern Iran. Endocr pract 24. Jawa A, Raza F, Qamar K, Jawad A, Akram J.
2005;11:313-8. Gestational diabetes mellitus is rare in
primigravida Pakistani women. Indian J
14. Ramirez T. Gestational diabetes mellitus; Endocrinol Metab. 2011;15:191-3.
Experience at a thrird level hospital. Gynecol
Obstet Mex 2005;73:484-91. 25. Rhee SY, Kim JY, Woo JT, Kim YS, Kim SH. Familial
clustering of type 2 diabetes in Korean women
15. Scott DA, Loveman E, Mclntyre L, WAaugh N. with gestational diabetes mellitus. Korean J Intern
Screening for gestational diabetes: a systematic Med 2010;25:269-72.
review and economic evaluation. Health Technol
Assess 2002;6:1-172. 26. Catherine KIM, Tiebin LIU, Gloria L. Doe’s frank
diabetes in first-degree relatives of a pregnant
16. Coustan DR, Nelson C, Carpenter. Maternal age woman affects the likelihood of her developing
and screening for gestational diabetes: a gestational diabetes mellitus or non-gestational
population based study. Obstet Gynecol, diabetes? Am J Obstet Gynecol
1989;73:557-561. 2009;201:576.e1–576.e6.

17. Hughes PF, Agarwal M, Newman P, Morrison J. 27. Campbell N, Pyke DA, Taylor KW. Oral glucose
Screening for gestational diabetes in multi- tolerance test in pregnant women with potential
ethnic population. Diabetes Res Clin Pract diabetes, latent diabetes and glycosuria. J
1995;28:73-78. Obstetric and Gyrtaecol Brit Comm Wealth
1971;78:498-504.
18. Green JR, Pawson IG, Schumacer LB, Perry J.
Glucose tolerance in pregnancy: Ethnic variation 28. A. van Leeuwen M, Zweers EJ, Opmeer BC, van
and influence of body habitus. Am J Obstet Ballegooie E, ter Brugge HG, de Valk HW, et al.
Gynecol, 1990, 163: 86-92. Comparison of accuracy measures of two
19. Granat M, Sharf M, Copper A. Glucose intolerance screening tests for gestational diabetes mellitus.
during pregnancy. Obstet Gynecol 1979, 53:157- Diabetes Care 2007;30:2779–84.
161.
29. Rutjes AW, Reitsma JB, Di Nisio M, Smidt N, van Rijn
20. Cypryk K, Pertynska MM, Szymczak W, Zawodniak JC, Bossuyt PM. Evidence of bias and variation in
SM, Wilczynski J, Lewinski A. Overweight and diagnostic accuracy studies. CMAJ
obesity as a common risk factor for gestational 2006;174:469–76.
diabetes, perinatal macrosomy in offspring and

Professional Med J 2014;21(2): 360-366. www.theprofesional.com 366