Objective
The objective of the study was to determine whether the alveolar-arterial (A-a) oxygen gradient is an adequate screening test for pulmonary embolism (PE) in pregnancy and postpartum.
Study Design
A chart review was performed at Tampa General Hospital. Patients who had a workup for a PE consisting of a computed tomography pulmonary angiogram and an arterial blood gas from 2002 to 2009 were included in the analysis. Sensitivity, specificity, and negative and positive predictive values were calculated. Additionally, common clinical signs and symptoms were assessed for their ability to accurately predict PE.
Results
Of 102 patients, there were 13 PEs (2 antepartum and 11 postpartum). The best sensitivity, specificity, and negative and positive predictive values for A-a gradients were 76.9%, 20.2%, 80.0%, and 11.5%, respectively.
Conclusion
The A-a gradient is a poor screening test for PE in pregnancy and postpartum. Suspicion of PE should prompt early imaging studies to rapidly make the diagnosis and begin treatment.
Pulmonary embolism (PE) is a rare but life-threatening condition in pregnancy. In the developed world, PE is the leading cause of maternal death. The incidence of PE has been estimated to occur in 1 of 1000 to 1 of 2000 pregnancies. Early diagnosis and treatment of PE are essential to minimizing maternal and fetal morbidity and mortality.
The gold standard for diagnosis of PE in nonpregnant patients is computed tomography pulmonary angiogram (CTPA). There is controversy over the best test in pregnancy; however, CTPA is considered by many to be the first-line test in pregnancy as well.
The biggest concern for performing a diagnostic radiologic test in the pregnant woman is radiation exposure to the fetus and mother. Fetal radiation exposure from CTPA in the third trimester can range from 60 to 230 μGy but can be further reduced by up to 50% with fetal radiation reduction strategies. Ideally, a screening test should determine patients at highest risk for PE and guide management regarding imaging studies to avoid exposing mother and baby to unnecessary radiation. One of the most common screening tests for PE is the alveolar-arterial (A-a) oxygen gradient, determined from an arterial blood gas (ABG).
Studies from the nonpregnant population have had mixed results with regard to the efficacy of the A-a gradient as a screening test for PE. Normal A-a gradients in women determined to have a PE have been reported as low as 1.9% to as high as 14%, with a wide range of positive and negative predictive values. Despite this controversy over the efficacy of A-a gradient, many providers utilize it as a first-line test in pregnant or postpartum patients when screening for PE.
We hypothesized that performing an ABG and A-a gradient is a poor screening test for the workup of PE in pregnancy and postpartum.
Materials and Methods
We conducted a retrospective study of patients admitted to Tampa General Hospital between June 2002 and June 2009. At our institution, CTPA is the gold standard test for diagnosis or exclusion of PE. It is our policy to perform a CTPA in pregnant or postpartum patients with signs and/or symptoms suspicious for PE.
Patients were selected for inclusion in the study by determining all patients either pregnant or up to 6 weeks postpartum who had a CTPA performed for a workup of suspicion for PE. Pregnant patients who had a CTPA performed for reasons other than suspicion for PE were excluded. Additionally, all patients had to have an ABG drawn prior to the CTPA to be included in the analysis. We excluded patients with a prepregnancy history of known cardiopulmonary disease.
The A-a gradient was calculated by the standard formula: A-a = [(P B – P H2O )(FiO 2 ) – (PaCO 2 )(1.25)] – PaO 2 . P B (mean barometric pressure) is 760 mm Hg, P H2O is 47 mm Hg, and FiO 2 is 0.21 for room air (adjusted if on O 2 ). We chose to use 3 variations of the normal value for A-a gradients based on previous studies: age/4 + 4 or less, 20 or less, and 15 or less. The calculated A-a gradient value was then compared with all 3 normal cutoff values.
Additionally, we sought to determine whether certain signs and symptoms typically associated with acute PE (tachycardia, tachypnea, hypotension, hypoxia, dyspnea, and chest pain) were helpful in screening for PE in pregnancy.
Sensitivities, specificities, negative predictive values (NPVs), and positive predictive values (PPVs) were calculated to determine the efficacy of the different A-a gradient parameters as screening tests for PE. A receiver operating characteristic (ROC) curve was created to evaluate the utility of the A-a gradient as a screening test for PE (SPSS version 18.0; SPSS, Chicago, IL).
This study was approved by the Office of the Institutional Review Board at the University of South Florida and Tampa General Hospital.
Results
Initially, 670 charts were reviewed (patients who had a CTPA and were pregnant or postpartum). One hundred two patients met inclusion criteria for analysis in the study (had an ABG and CTPA for workup of suspected PE). Maternal ages ranged from 15 to 44 years (mean of 27 ± 7 years). There were 13 PEs in the study population (12.7%). Of those with a PE, 2 were antepartum (15.4%, 19 and 32 weeks’ gestational age) and 11 were postpartum (84.6%). Table 1 demonstrates the sensitivity, specificity, PPVs, and NPVs for the 3 parameters of normal A-a gradients. The age-adjusted formula provided the highest sensitivity (76.9%), but the cutoff value of 20 demonstrated the highest specificity (20.2%) of the parameters tested.
| A-a gradient | Sensitivity, % | Specificity, % | PPV, % | NPV, % |
|---|---|---|---|---|
| Age adjusted a | 76.9 | 13.5 | 11.5 | 80.0 |
| >15 | 69.2 | 14.6 | 10.6 | 76.5 |
| >20 | 61.6 | 20.2 | 10.1 | 78.3 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
