Pregnant women with preexisting diabetes, particularly type 1 diabetes, face significant challenges. Optimal maternal, fetal, and neonatal outcomes are the goal of care for pregnant women with preexisting diabetes. In addition to the stresses and demands of any pregnant woman, one who has a pregnancy that is complicated by diabetes may need lifestyle changes, frequent appointments, intensive fetal surveillance, and rigorous management of diabetes to achieve tight glycemic control. However, women with diabetes often experience barriers to optimal care. These may include financial limitations, a poor support system, lack of knowledge about diabetes, and psychiatric comorbidities, as well as lack of access to the comprehensive care needed.1 The following case study illustrates a woman with a history of poor glycemic control.
LITERATURE REVIEW
Optimal pregnancy outcomes are achieved when the maternal glucose level is maintained within the reference range, avoiding levels that are below the reference range as well as those above.2 Glycemic control over time can be determined using the A1c test, also known as glycosylated hemoglobin test. Plasma glucose attaches to the hemoglobin in red blood cells (RBCs). The A1c test gives an estimate of glucose levels over the past 3 to 4 months as the life span of RBCs is approximately 120 days. Glycosylated hemoglobin results are based on a weighted mean; 50% of the result reflects the most recent month, and 25% the previous month and the other 25% of the value3 is based on the glycemic control in months 3 and 4. Risk for adverse outcomes increases when A1c values are elevated; this is on a continuum with risk increasing as the A1c value increases.2 The A1c range for a nondiabetic pregnancy4 is lower (4.4%-5.7%) than the range found in the nonpregnant woman (4.7%-6.3%). Preconception glycemic control and control in early pregnancy have a significant impact on the increased rates of congenital anomalies and spontaneous abortions that are witnessed in diabetic pregnant women.5
It is advised that any woman with diabetes between menarche and menopause be reminded of the importance of achieving euglycemia before conception.5 Counseling regarding using an effective contraceptive is needed until target A1c levels are established. It is advisable that when pregnancy is contemplated, the woman with preexisting diabetes consult with a team specializing in diabetes and pregnancy to ensure glycemic control and to identify complications of diabetes that need to be addressed prior to conception.1 Major congenital anomalies are seen at a rate of 6% to 9% in infants of women with preexisting diabetes, compared with the 2% to 3% rate in the general population.5 When the A1c rate is close to 10%, the rate of fetal anomalies is 20% to 25%.6,7,8
Poor glycemic control continues to increase risk for the fetus throughout pregnancy, even after organogenesis is complete (Table 1). Type 1 diabetes is associated with a 3 to 5 times increased risk for stillbirth.10 It has been postulated that this may be explained by the fetal hypoxia and acidosis that is caused by hyperglycemia.11 After 12 weeks' gestation, the fetal response to maternal hyperglycemia is hyperinsulinemia. This promotes excess fat storage, potentially resulting in macrosomia, defined as birth weight more than 4000 to 4500 g.12 This increases the risk of a number of newborn complications including shoulder dystocia, organomegaly, hypoglycemia, respiratory distress, electrolyte abnormalities, and hyperbilirubinemia.13 Infants of diabetic mothers with poor maternal glycemic control and ketonuria have an increased risk of neuropsychological deficits. These infants are also at risk for developing obesity and diabetes later in life,1 and continued poor glycemic control throughout pregnancy might compromise the child's intellectual and psychomotor development.2
Diabetic ketoacidosis (DKA) is estimated to occur in 1% to 2% of pregnancies complicated by preexisting diabetes.14 During pregnancy, DKA may occur at lower glucose levels than in nonpregnant women. Ten to thirty percent of DKA during pregnancy occurs in women who have glucose levels less than 250 mg/dL.15,16 DKA poses a significant risk for the fetus, with perinatal mortality rates reported between 9% and 35%.17 Factors that increase the risk of DKA in pregnant patients include infections (eg, urinary tract infections or influenza), stress, trauma, lack of patient compliance, and treatment with specific medications used in obstetrics including corticosteroids and [beta]-mimetic tocolytics such as terbutaline.15,17 It is also important when assessing pregnant women to be cognizant of the clinical manifestations of DKA such as abdominal pain, nausea and vomiting, and an altered level of consciousness.18
Hypoglycemia has been reported to cause maternal death.11 A study that was conducted did not demonstrate that hypoglycemic episodes in and of themselves cause fetal harm.19 However, recognition and management of hypoglycemia for patients receiving insulin are essential. Women's families ought to be educated regarding symptoms of hypoglycemia (see Table 2) and the appropriate treatment. It is recommended that diabetic women be advised of the importance of carrying a form of glucose and wearing some form of medical alert identification.2 It is also suggested that all pregnant women with preexisting diabetes, particularly women with type 1 diabetes, be given a prescription for a glucagon kit, which can be used if the woman is unable to self-treat. It is important for the woman to identify the significant others who can be educated to recognize when glucagon is needed and how to give the injection. This may be accomplished by teaching them to give the woman a subcutaneous injection with normal saline and reviewing use of the glucagon kit.19 For nonpregnant patients, 72 mg/dL is considered a conservative threshold in which to treat hypoglycemia.20 In pregnancy, the threshold is lower at 60 mg/dL.22 The risk of hypoglycemia increases during pregnancy owing to fetal glucose consumption, a decreased maternal response to counterregulatory hormones, and intensive glycemic management.1
Women with type 1 diabetes, owing to a diminished epinephrine response to hypoglycemia and decreased autonomic symptoms, may develop hypoglycemia unawareness.20 Symptoms may not occur until glucose levels are quite low; therefore, advise patients to monitor blood glucose levels frequently and have a glucagon kit readily available.20
Pregnant women with preexisting diabetes have an increased incidence of hypertensive disorders including chronic hypertension, preeclampsia, chronic hypertension with superimposed preeclampsia, and gestational hypertension.23 Other maternal complications are increased rates of hydramnios and preterm delivery.24
Pregnancy has not been shown to cause a progression of microvascular complications.25 However, proliferative retinopathy may progress during pregnancy if not treated with laser photocoagulation prior to conception.26
According to the American College of Obstetrics and Gynecology, women with pregnancies that are complicated by type 1 diabetes require intensive management; prenatal visits may need to occur every 1 to 2 weeks until 28 weeks' gestation. After that time, weekly visits are recommended.27 The American Diabetes Association (ADA) published consensus panel recommendations for the management of pregnant women with preexisting diabetes on the basis of a thorough review of the evidence.11 Patient-centered multidisciplinary care is advocated by the ADA consensus panel owing to the many complex issues experienced by this population. Multidisciplinary team members may include perinatologists, obstetricians, other medical providers, nurses, certified diabetes educators, dieticians, and social workers. Referrals for specialty care, for example, to an ophthalmologist, podiatrist, or psychiatrist, may be made based on assessments of the patient's condition.11
Psychosocial assessment and intervention is an essential component of patient-centered team management. In discussing the role of multidisciplinary care, the ADA consensus panel believes that the clinician who is caring for the pregnant diabetic woman must walk a fine line between the tension generated by focusing on emotional distress versus concentrating on the practical issues of diabetes management.11 For example, depression during pregnancy is associated with poor self-care.28 Management of pregnancy that is complicated by diabetes requires intense management from the multidisciplinary team; however, it is the woman who has the greatest work to do in achieving a healthy pregnancy. Self-monitoring of blood glucose levels (as many as 10 times per day), multiple daily injections or management of an insulin pump, strict dietary adherence, and numerous medical appointments are among the challenges for patients.11 These challenges necessitate assessment of the woman's ability to cope with the plan of care. The ADA consensus panel recommends the use of a psychosocial screening tool developed by the American College of Obstetrics and Gynecology and modified for use in assessing pregnant diabetic patients. On the basis of results of the assessment, intervention and referral to mental health providers may be arranged. The tool can be found in Management of Preexisting Diabetes and Pregnancy, the ADA's consensus panel publication.11
CASE PRESENTATION
A 26 year old, gravida 3, para 0111, at 35 2/7 weeks' gestation, (height 5 ft 4 in, weight 130 lb) was transported to a tertiary academic medical center from an outlying facility. The father of the infant found the woman unresponsive in her home 4 hours prior to arrival at the tertiary center. Using the woman's glucometer, the significant other found that the capillary blood glucose level of the woman was 13 mg/dL and immediately called 911. Glucagon was given intravenously by a paramedic en route to the referring hospital, but the woman did not regain consciousness. The woman had 2 seizures en route to the referring hospital and had 3 additional seizures in their emergency department. At the referring hospital, the woman received Ativan and was placed on the paralytic agent vecuronium. The referring hospital intubated the woman for airway management and initiated treatment with magnesium sulfate for seizure prophylaxis. A decision was made to transfer the woman to a tertiary level center for more intensive care. During transport to the tertiary center, the woman had an additional seizure and was again given Ativan.
The patient's medical history was significant for type 1 diabetes, White's classification D (onset less than 10 years of age or duration greater than 20 years with positive vascular disease) diagnosed at age 7 with a history of poor glycemic control. Five days prior to this admission, the woman was admitted for DKA at a third hospital, the one at which she received prenatal care. Blood glucose values during that admission ranged from 19 mg/dL to 529 mg/dL. At that time, the woman's insulin regimen was changed to 18 units of regular and 32 units of NPH in the AM and 10 units of regular and 20 units of NPH in the PM. Of significance is that the woman signed herself out of the hospital against medical advice 3 days prior to this admission at the tertiary center. The surgical history was significant for a cesarean section performed for a nonreassuring fetal heart rate tracing 3 years prior to this pregnancy. A review of the patient's prenatal record indicated that the prenatal laboratory test reports were within normal limits (see Table 3)
The woman's family, including her mother and the father of the infant, were present at the time of admission to the tertiary center and provided valuable information about the woman's difficulties in the management of her diabetes. The family reported that the patient was not compliant with the diabetes regimen including diet and insulin management. It was reported that there were frequent episodes of severe hypoglycemia (with symptoms that included bowel and bladder incontinence and posturing) requiring glucagon administration. The family also noted that there had been numerous prior admissions for DKA. The significant other also revealed that an argument had occurred the afternoon before the woman was found unresponsive and that during the argument, the woman had threatened to take an overdose of insulin.
Admission vital signs at the tertiary center were as follows: blood pressure, 179/100; pulse, 137; respirations, 12; and temperature, 98.0[degrees]F. The ventilator settings were SIMV, FIO2, 50%; tidal volume (VT) 500; rate, 12; and PEEP (positive end expiratory pressure), 5. The woman was nonresponsive to either verbal or deep pain stimuli. The woman's gag and corneal reflexes were both intact and pupils were equal, though with a sluggish response. The rest of the physical examination was unremarkable; lungs were clear to auscultation bilaterally and the abdomen was soft with a fundal height of 34 cm. The initial capillary fingerstick blood glucose level was 252 mg/dL, and laboratory test results did not indicate the presence of preeclampsia. The hemoglobin and hematocrit (Hgb/Hct) count values were 8.5 g/dL and 25.7 %, respectively. The liver function tests were normal; however, serum creatinine level was elevated to 1.16 mg/dL, and the initial serum blood glucose level was 192 mg/dL. A urinalysis was positive for an active urinary tract infection, and intravenous antibiotic therapy was initiated (see Table 4). An A1c test was not performed during this admission; however, a result from 4 days prior was 8.6%. An A1c of 9% correlates with a mean plasma glucose level3 of 240 mg/dL. The woman was transfused with 2 units of packed RBCs after admission to the tertiary center.
Peripheral venous access was established by the paramedic prior to the woman's arrival at the referring hospital. A second peripheral IV, in addition to a radial arterial line, was placed upon arrival at the tertiary center. Clear amber urine was draining via a Foley catheter in amounts greater than 30 mL/h.
Continuous electronic fetal and uterine monitoring was initiated; the fetal heart rate baseline was 150 beats per minute with absent variability, no accelerations, and no decelerations (see Figure 1). Uterine activity was noted, with contractions every 1 to 2 minutes, mild to palpation, lasting 40 to 60 seconds with soft uterine tone between contractions. There was no observed response by the woman to the contractions. Initial vaginal examination was fingertip dilated, 70% effaced, high station, cervix mid position and soft. Once preeclampsia was ruled out, a decision was made to continue the treatment with magnesium sulfate at 2 g/h for tocolysis.
Upon arrival at the tertiary center, the vecuronium treatment was discontinued in an effort to further evaluate the neurologic status of the woman and the fetus. The maternal-fetal medicine (MFM) specialists felt that the vecuronium could be masking a positive change in the condition of both the patient and the fetus owing to its sedative properties. The patient's initial Glasgow Coma Scale score was 3 out of 15. As the woman's fingerstick blood glucose levels continued to be elevated after admission (213-252) the woman was started on an intravenous infusion of regular insulin at 1 unit per hour, approximately 4 hours after arrival.
Consultations were requested from neurologists, anesthesiologists, and neonatologists. The team of neurologists recommended obtaining a magnetic resonance arteriogram and a magnetic resonance venogram to evaluate the patient's brain for any abnormalities. Physicians and nurses worked with the family in an effort to keep them informed of the plan of care and the woman's clinical condition and to include them in decision making. Pastoral care met with the family to provide additional support.
The woman required intensive care including continuous assessment to observe for any changes in status, to monitor the elevated blood pressures (140/100-170/90 mm Hg) and the extremely labile capillary blood glucose values, which ranged from 134 mg/dL to 287 mg/dL. Both the diabetic coma and the seizure activity put the fetus at significant risk for brain damage secondary to a lack of oxygenation. Approximately 4 hours after admission the fetal assessment findings were as follows: absent fetal heart rate variability, a biophysical profile of 2 out of 10, and occasional late decelerations (see Figure 2).
These clinical findings were indicative of a potential severe neurologic outcome for the infant. In most situations, this would warrant an immediate cesarean section. However, the MFM physicians felt that given the critical and unstable condition of the woman, a cesarean section could lead to further stress and possibly jeopardize the woman's life and long-term outcomes. The MFM physicians discussed this thoroughly with the family, who agreed with the decision.
The magnetic resonance arteriogram and the magnetic resonance venogram reports indicated no evidence of thrombosis or clot, and an electroencephalogram (EEG) report revealed diffuse cerebral dysfunction, which might indicate hypoxic/ischemic encephalopathy, toxic/metabolic disorders, or a postictal state. In this case, although the initial blood glucose level when the woman was found unresponsive was 13 mg/dL, the length of time of severe hypoglycemia was not determinable; the woman also experienced multiple seizures. The EEG report noted that there was no brain reactivity with both verbal or painful stimulation and specifically mentioned excess delta activity, potentially reflecting the effects of hypoglycemia on the brain. The initial chest radiograph was normal.
Through the initial 24 hours of admission, the neurologic status did not change. Neither verbal nor deep pain stimuli elicited response, but occasional spontaneous movements of the woman's arms, legs, and feet were noted. A repeat EEG indicated no improvement with continued generalized diffuse slowing and high voltage in the delta range. Ventilator settings were decreased; however, the woman did not tolerate extubation. Blood pressures remained labile, ranging from 125/70 to 185/95 mm Hg, which was concerning as there was no history of hypertensive disease. The elevated blood pressure levels were managed with intravenous hydralazine initially and then intravenous labetalol. The goal was to maintain the systolic blood pressure below 160 mm Hg.
The fetal heart rate status did improve slightly after 12 hours as evidenced by minimal and occasionally moderate variability with no decelerations or accelerations (see Figure 3). Contractions continued every 2 to 6 minutes although the patient remained on IV magnesium sulfate treatment. Several vaginal examinations throughout this time revealed no evidence of cervical change. A repeat biophysical profile also indicated improvement in the fetal status with a score of 6 out of 10 (minus 2 for tone and 2 for a nonreactive nonstress test).
Thirty-seven hours after admission, with no change in maternal neurologic status but with blood glucose level and blood pressure stable, a decision was made to proceed to delivery via repeat cesarean section. To reach this decision, a meeting was held consisting of the woman's mother, the father of the infant, the MFM attending, and a neonatologist. As with the initial meeting, the family was once again counseled that there was a high likelihood of fetal brain injury secondary to severe maternal hypoglycemia and multiple maternal seizures. The woman's family verbalized understanding of the potential neonatal outcomes and agreed to proceed with an operative delivery for maternal indications. Given that the woman had stabilized, it was felt that the woman would have the best opportunity for improvement once delivered. During this time, the woman's blood glucose values continued to fluctuate somewhat and the insulin infusion was titrated in an attempt to maintain a blood glucose value below 120 mm Hg. The woman was transfused with an additional 2 units of packed RBCs for a hematocrit level of 21.8% prior to delivery. There was no potential source of bleeding identified, and the cause of anemia was not determined.
A male newborn was delivered via repeat cesarean section under general anesthesia with Apgar scores of 8 and 9, at 1 and 5 minutes, respectively. The umbilical artery pH was 7.36 and the umbilical vein pH was 7.41, which are within the reference range. The infant was transferred to the neonatal intensive care unit for close monitoring; he transitioned well and required no supplemental oxygen. The infant's birth weight was 2205 g, which was appropriate for gestational age. The infant's initial heelstick glucose level was 71 mg/dL. Intravenous access was initiated with dextrose 10% in water (D10W) to provide hydration and as a prophylactic measure against neonatal hypoglycemia. The infant's blood glucose values remained within reference range and no additional interventions were required. The infant was observed closely for any potential hypoxia-related sequelae, including necrotizing enterocolitis, and hypoxic-ischemic encephalopathy. As there was a concern about the infant developing necrotizing enterocolitis, oral feedings were withheld and total parenteral nutrition was initiated. Phototherapy was started on the third day of life for hyperbilirubinemia caused by prematurity. The infant's maximum bilirubin level was 12.7 mg/dL. After oral feedings were initiated, it was noted that the infant had feeding difficulties and gastroesophageal reflux. At approximately 1 month of age, the infant was transferred to a pediatric continuing care hospital where a feeding progression program continued. To date, the infant is at home with family and is reported to be doing well.
Postoperatively, there continued to be no change in the woman's neurologic status. Consequently, the woman was transferred to the medical intensive care unit for ventilator management and neurologic monitoring. The woman was extubated on the second postoperative day and placed on a venturi mask at 40% FIO2. A computed tomography scan revealed no acute intracranial pathologic abnormalities and no evidence of infarction. After a week in the medical intensive care unit, there was minimal improvement in neurologic status. The woman continued to have spontaneous uncoordinated movements of both arms and legs. There was no response to verbal commands; however, spontaneous opening and closing of the eyes and an occasional response to deep pain stimuli began occurring. Opinions from the team of neurologists regarding the long-term prognosis ranged from potential significant improvement with minimal assistance needed for activities of daily living, to a possibility of minimal improvement resulting in a need for full-time care.
On day 14, the woman was transferred to a long-term care facility with a coma emergence program, where significant progress in neurologic status was noted. The woman began to walk, respond to verbal commands, and engage in conversations. Once the woman regained the ability to swallow and eat, diabetic management became difficult; other patients gave her food and she would take food from their trays as well. These behaviors contributed to numerous episodes of both hypoglycemia and DKA. The woman remained at this facility for approximately 41/2 months before returning to the family home. She is able to perform self-care with assistance. While the patient does have some communication and cognitive deficits, there are no obvious motor deficits.
CASE ANALYSIS
When caring for a pregnant diabetic woman admitted with complications, a thorough assessment includes past medical and surgical history, a physical examination, and psychosocial issues. The current insulin regimen, including carbohydrate to insulin ratios or the insulin pump's basal and bolus settings, needs to be determined. The prescribed nutrition plan and adherence to diet as well as insulin administration and self-monitoring of blood glucose levels are assessed. The history of diabetes-related complications is ascertained. Examples of diabetes-related problems that may complicate the woman's condition include hypertension, kidney disease, retinopathy, gastroparesis, and heart disease. If the woman has a history of hypertension or other vascular complications, there is an increased risk for preeclampsia.20
Assessments for the current glucose value, the presence of ketones, and any physical or laboratory findings that indicate the woman has or is at risk for DKA ought to be performed. It is recommended that, as part of the thorough physical examination, a careful assessment of the patient's skin for any breakdown or open wounds be performed. An assessment of the woman's attitude about diabetes and the pregnancy, social support system, and coping skills as well as the presence of any psychosocial problems that indicate a need for referral to the social worker must also be completed. In addition, it is suggested that the patient's knowledge of diabetes self-care and self-care behaviors be assessed. Do not assume that there is an adequate knowledge base no matter how many years since the diagnosis of diabetes.9
Reviewing the woman's blood glucose level logs and A1c results will give valuable information about the patient's past glycemic control, including any history of hypoglycemic episodes and symptoms, or lack of awareness. Women who experience frequent episodes of hypoglycemia may have hypoglycemia unawareness. If this occurs, the woman may reach extremely low glucose level without symptoms and may need careful monitoring.20 This actually seemed to be the case with this woman. During the woman's prolonged stay at the long-term care facility, it was noted on multiple occasions that when blood glucose levels were as low as 35 to 40 mg/dL frequently no symptoms of hypoglycemia were seen. However, when the woman's blood glucose level fell below 30 mg/dL she became symptomatic and required glucagon or other intense measures to reverse the hypoglycemia.
Interpretation of the EEG results require understanding that hypoglycemia may be the cause of neuronal death demonstrated by a flat EEG. This has been shown to occur when glucose levels fall below 1mM (18 mg/dL) and is dependent upon the body glycogen reserves.29 As the blood glucose levels progressively drop to the range of 1 mM to 2 mM, theta waves increase and coarse delta waves appear.29
Fetal resuscitation could be enhanced by adequate maternal resuscitation and stabilization.30 Absent fetal heart rate variability may indicate the presence of fetal hypoxia and/or acidosis.31 When interpreting abnormal fetal heart rate tracings, it is important to consider all possible causes of absent variability, such as maternal hypoglycemia or hyperglycemia, paralytic or sedative medications, anomalies, fetal heart abnormalities, hypoxia, and acidosis.30 The entire clinical profile of the mother and risks versus benefits to her health should be considered when planning interventions to improve the fetal status.
Staffing is a primary concern when caring for a critical patient. This woman initially required 2 nurses to assess, admit, and provide care. An experienced nurse is able to anticipate the needs of these women with complex problems and provide direction and guidance to others involved in delivering care. Nursing interventions are based on assessment, findings, and outcomes. Ancillary staff members are necessary to provide additional assistance such as gathering equipment and supplies. Involve the respiratory therapy department to set up a ventilator in the labor and delivery unit and monitor the ventilator settings on the basis of the woman's condition. An anesthesiologic consult is necessary immediately in the event an emergent delivery is required, if central venous or arterial access is needed, and/or to manage airway issues. Maternal-fetal medicine physicians, if not already available, are contacted immediately upon notification of the woman's acceptance for transfer. It is essential to have a MFM specialist evaluate the problem as soon as possible upon arrival. Notify the neonatal physicians and the neonatal team as quickly as possible. This will allow the team preparation time for the arrival of a critically ill infant and to be available to consult with the family about potential neonatal issues. Additional consults including those with neurologists and endocrinologists may be necessary.
A lack of consistent care contributed to the outcome of this woman's case. Many different providers saw the woman at multiple hospitals. There appeared to be little to no communication among the providers regarding the woman's condition and the changes in therapy that occurred. According to the prenatal record, the woman began prenatal care only approximately 3 months prior to the delivery. This as well as the woman's noncompliance with the care regimen likely precipitated or at least contributed to the severity of the hypoglycemic episode.
SUMMARY
This pregnant woman endured a prolonged hospitalization and has permanent deficits that will make it difficult to live independently, care for children, and manage the diabetic regimen. While the woman received comprehensive intensive care and extensive rehabilitation that minimized long-term sequelae, the focus should be on preventing this type of incident. During the prenatal period, this woman would have benefited from psychological screening and intervention as well as intensive diabetes management. This would include the development of a comprehensive plan of care. If the woman has a history of hypoglycemia unawareness, the plan should include specific instructions regarding monitoring and intervention. When care is consistent, patient centered, and multidisciplinary, the type of incident that was reported in this case presentation may be avoided. Nurses can play an essential role in facilitating the team approach in the care of pregnant women with preexisting diabetes and assuring holistic and comprehensive care.
REFERENCES