The Impact of Pregnancy on the Evaluation of Chest Pain and Shortness of Breath in the Emergency Department

Abstract

Background

Pregnant patients are at an increased risk for severe morbidity and mortality when compared to the general population. Imaging studies have become an integral part of the work-up for patients who present to the emergency department. However, lack of clear guideline has led to confusion regarding the appropriate indications and protocols for examining pregnant patients.

Objective

To investigate the comparative rates of imaging studies between pregnant and non-pregnant female patients who presented for emergent evaluation of chest pain and shortness of breath.

Study Design

All reproductive age females presenting to the emergency department at an academic teaching institution with a chief complaint of “chest pain” or “shortness of breath” from 2010 to 2015 were identifi ed. Cohorts were divided based on pregnancy status and chief complaint. Utilization rates of imaging studies were compared between cohorts.

Results

Over the study period 4,834 women were included. One hundred and seventy-four were pregnant. Pregnant patients with “chest pain” or “shortness of breath” were signifi cantly more likely to undergo a venous duplex, but less likely to undergo a chest X-ray as compared to non-pregnant patients. There was no difference in the rates of chest computed tomography (CT) imaging or magnetic resonance imaging (MRI) based on pregnancy status in our data set.

Conclusion

Imaging studies are an integral adjunct for evaluation in patients reporting “chest pain” and “shortness of breath.” Pregnancy places patients at an increased risk of severe sequelae requiring prompt diagnosis to prevent harm to the mother and fetus. American College of Obstetrics and Gynecology (ACOG) Committee Opinion 656 clearly states that, with few exceptions, radiography should not be withheld from pregnant patients. This study suggests that pregnant patients are signifi cantly less likely to undergo radiography, which could place them at increased risk for delayed diagnosis and treatment.

Introduction

Pregnancy predisposes women to anatomic and physiologic changes that result in an increase in the incidence of conditions that may be life threatening to patients in this subpopulation [1]. Causes of chest pain or shortness of breath (SOB) in pregnant patients can often be attributed to anatomic and physiologic changes. SOB is common during pregnancy and occurs in 60% to 70% of healthy pregnant women [2,3]. Nevertheless, such complaints require investigation to rule out pathologic processes that may jeopardize maternal and fetal health. The differential diagnoses for SOB and chest pain in pregnantwomen vary. They include asthma, valvular heart disease, pulmonary edema, peripartum cardiomyopathy, aortic dissection, musculoskeletal issues, gastrointestinal reflux disease (GERD) and acute coronary syndrome. Moreover, life threatening conditions such as pneumomediastinum, pneumothorax, pneumonia and pulmonary embolism require prompt diagnosis and management to save both mother and fetus [4-11].

Over the past two decades, there has been an increase in the use of radiologic examinations in pregnant women [6-8]. High radiation doses in pregnancy can result in growth restriction, microcephaly and potential fetal loss. Within the medical community,however, there are demonstrated gaps in the knowledge of providers regarding radiation in the pregnant population [9,10].

In 2016, the American College of Obstetrics and Gynecology (ACOG) released a committee opinion centered on recommendations and guidelines for diagnostic imaging during pregnancy. Diagnostic imaging should be performed during pregnancy with an understanding of the maternal and fetal risks as well as benefits. These modalities should not be withheld where indicated or necessary. Ultrasonography and magnetic resonance imaging (MRI) are not associated with radiation risk. As such, they are the imaging techniques of choice for the pregnant patient but should be used prudently and only when use is expected to answer a clinical question or has medical benefit. This committee opinion, among other studies, has shown that the risk from radiation depends on the dose and gestational age [1,12-16]. Moreover, the radiation dose from a single diagnostic imaging is highly unlikely to exceed the estimated threshold dose of 100 mGy for the inductionof malformations and carcinogenesis [1,6-8,16,17]. As a result, ACOG does note that radiation exposure through radiography, computed tomography (CT) scan, or nuclear medicine imaging is at a dose lower than the exposure associated with fetal harm [17]. Basedon this, ACOG states withholding imaging studies due only to pregnancy status is not indicated.

Misconceptions about maternal and fetal risk of radiation exposure have led to the conservative use of these imaging modalities despite ACOG recommendations. In an attempt to evaluate and investigate complaints related to chest pain or SOB in pregnantpatients, imaging may be necessary to develop an accurate diagnosis. There remains a lack of clear emergency department guidelines regarding the use of radiologic imaging studies in the pregnant population. This drives many clinicians to proceed cautiously and limit the use of imaging modalities that could prove beneficial in the diagnosis and timely management of life threatening conditions in pregnant women. Furthermore, pregnant patients often question the potential effects of the radiation exposure and may perceivethe teratogenic risk of the test as high [9,18-22].

Our objective in this article is to investigate the comparative rates of imaging studies between pregnant and non-pregnant female patients who presented for emergent evaluation of “chest pain” and “SOB.”

Methods

This retrospective cohort study examined female patients who presented to the emergency department of an academic center with dyspnea or angina. It was approved by the institutional review board of the Pennsylvania State University (Study3466).

In this study data on adult females of reproductive age who presented to the emergency department with a chief complaint identified in the medical record as “chest pain” or “SOB” was collected. Records were collected on emergency room visits from 2010 to 2015. Reproductive age was defined as 18–52. Patients were excluded if there was a previously diagnosed venous thrombosis or if the patients left the emergency department against medical advice prior to completion of the evaluation. Cohorts were defined based on chief complaint and further divided by pregnancy status.

An electronic medical record (FirstNet) associated specifically with the emergency department was utilized for record collection. Chief complaints were placed into this database by the department’s intake nurse. Patient tests and imaging studies wereidentified by extracting the provider orders which are uniformly inserted electronically. The discharge disposition of a patient from the emergency departmentis likewise tracked by the provider in the medicalrecord. Patient characteristics, evaluation and dispositionwere abstracted by means of predefined criteria.Statistical analysis was performed using SAS9.4. For measuring binary outcomes logistic regressionadjusting for age at the visit was utilized forcalculating odds ratios. Continuous variables werecompared with quantile regression adjusted for age atpresentation.

Results

The overall cohort comprised 4,834 patients.Three thousand nine hundred and fifty-four patientspresented with angina symptoms and were classifiedwith a chief complaint of “chest pain.” Of these 109 were pregnant at the time of evaluation. Patients whowere not pregnant were significantly older; there wasno significant difference in race or insurance status (Table 1).

Table 1. Comparative rates of imaging studies in the emergency department adjusted for age.

CT: computed tomography; MRI: magnetic resonance imaging.

	Chest pain			Shortness of breath
	Pregnant (n = 109) n (%)	Non-pregnant (n = 3,845) n (%)	p value	Pregnant (n = 65) n (%)	Non-pregnant (n = 815) n (%)	p value
Venous duplex	10 (9.2)	59 (1.5)	< 0.001	6 (9.2)	22 (2.7)	0.002
Chest CT	24 (22.0)	906 (23.6)	0.813	19 (29.2)	208 (25.5)	0.222
Chest X-ray	57 (52.3)	3,267 (85.0)	< 0.001	17 (26.2)	627 (76.9)	< 0.001
MRI	2 (1.8)	29 (0.8)	0.131	0 (0.0)	0 (0.0)	—

Examination of demographic data for patientswith angina symptoms revealed non-pregnant patientswere significantly more likely to be older (p < 0.001) and have private insurance (p = 0.013). Pregnant patientswere more often covered by Medicaid. Therewas no difference in race between the cohorts. Therewas no difference noted in the likelihood of patientshaving a primary care provider (p = 0.039).

As part of the analysis all variables were adjustedto account for the difference in age between cohorts. With regards to radiologic imaging there was nosignificant difference in frequency of CT (p = 0.813) or MRI (p = 0.131) studies. Pregnant patients had asignificantly higher likelihood of having a venous duplexperformed as part of their evaluation (p < 0.001). Chest radiographs were less likely to be performed ifa patient was pregnant (p < 0.001). Length of time betweenordering and performance of all imaging studieswas not influenced by pregnancy status. There wasno change in length of evaluation in the emergencydepartment between groups.

Of the 880 patients who presented to the emergencydepartment with dyspnea, coded as “SOB,” 65 were pregnant. In a similar finding to the patients withangina non-pregnant patients were significantly older (p < 0.001). There was no difference in race, or presenceof a primary care provider. Unlike patients presentingwith angina, pregnant patients with dyspneawere more likely to have private insurance comparedto their non-pregnant counterparts (p = 0.012).

After correcting for age, there was no differencein the likelihood of CT imaging between cohorts (p= 0.222). The pregnant population underwent morevenous duplex imaging (p = 0.002). Pregnant dyspneicpatients were less to likely to receive chest radiographsas part of their evaluation in the emergencydepartment (p < 0.001). There was no difference inlength of stay or time between ordering and performanceof imaging studies between cohorts.

Discussion

The emergency department is the location for the initial evaluation of patients presenting with simple and complex conditions. Evaluation of pregnant women in this setting elicits some level of hesitancy especially when radiographic imaging is warranted. The objective of our study was to review the comparative rates of imaging studies between pregnant andnon-pregnant female patients who presented for emergent evaluation of “chest pain” and “SOB.” After review of the literature, this is the first study of its kind. Our retrospective analyses support studies that have highlighted the trepidation associated with radiologicimaging in the pregnant population.

There is a discrepancy in the evaluation of pregnant verses non-pregnant patients in respect to ordering chest X-rays. Pregnant patients who presented with SOB and chest pain were less likely to receive chest X-rays as part of their evaluation in the emergencydepartment. This discrepancy may be due to a series of factors, uncovered by other studies, which point to concerns about radiologic imaging especially during pregnancy. Ntusi et al. report in their study that diagnostic X-rays and nuclear procedures emergeas the greatest source of concern, however, most diagnostic radiologic procedures do not expose the pregnant woman to a degree of radiation that would threaten the well-being of the developing pre-embryo, embryo or fetus [12]. The average estimated dose to the fetus from a single chest radiograph is 0.0005–0.0100 mGy [1,10,12,17]. This amount is significantly below the threshold of radiation dose that is deleterious to the fetus. The upper limit of radiation dose that is associated with growth restriction, central nervous system defects, malformations, carcinogenesis and intellectual impairment is 100 mGy. Some may even argue that the threshold dose is as low as 50 mGy [17,23]. Based on these studies, the average estimated fetal dose is significantly below the radiation threshold dose.

In an attempt to answer this question, studies have alluded to the possibility of knowledge gap that exists in physicians and residents about radiation risk and dose in patients [18,20-22]. In our study, we saw an increased likelihood for pregnant women to receive venous duplex as part of their evaluation as compared to non-pregnant women. This brings into question whether this knowledge deficit influences physicians to adopt alternative imaging techniques over others that are perceived as harmful to the mother and fetus. There is also a need to consider how the knowledge gap affects clinical management. Without guidelines, physicians may use their own discretion in selecting imaging modalities in the pregnant population. Studies support that they select a modality which isdeemed as safest in order to avoid radiation exposure [17]. Consideration and utilization of safer imagingmodalities in the initial evaluation process is prudent,however, this might come at a cost of delaying diagnosesand swift management.

This study did not reveal any significant differencein the frequency of CT or MRI usage inpregnant vs. non-pregnant patients. Compared to CT, chest X-ray was less likely to be used as a primaryimaging modality. While studies have reported thatchest radiograph and CT chest expose the fetus tosimilar amounts of radiation [17], this study shows thatphysicians are less likely to order chest X-rays. Thefetal dose from a single chest X-ray is 0.0005–0.0100mGy. CT chest delivers a fetal radiation dose of0.01–0.66 mGy [17]. This trend is not all that surprisingdue to reports of increase in the number of CT scansordered over the past three decades [7,24]. Lazarus etal., were able to obtain results, which showed an increasedutilization of CT pulmonary angiography inpregnant patient population [7]. Based on these results,we can safely speculate that this study’s findings supportthis trend. This global change in practice is likely secondary to the increased clinical suspicion of severe disease processes in this patient population, and accordinglya need for more definitive testing. There are several strengths to this study. This retrospective cohort study highlights an importantarea where there is a lack of research on pregnantpopulations. This study is the largest that we areaware of that examines imaging discrepancies in apregnant population. Our study had several importantlimitations, as this study is a retrospective chart review from a single academic medical center we cannotinfer causation from the associations we observed. Additionally, the results of this study to some degreemay not be generalizable to other institutions. We did not account and correct for physician preferences inordering specific imaging studies in the emergency department. Finally, we were not able to determinethe final diagnosis that patients received at discharge. It is possible that a baseline difference in diseaseprocesses between the cohorts could contribute to thedifferences seen.

In conclusion, this study highlights a discrepancyin the evaluation of pregnant and non-pregnantpatients in an emergent setting. Despite recommendationsto the contrary pregnant patients were significantlyless like to receive certain radiographicimaging specifically chest radiographs. Future effortsto promote the new recommendations and to improveknowledge about radiation exposure in the pregnantpopulation are essential to address this issue in theemergency department. Previous research suggeststhat additional education and training of health careproviders could limit this difference.

Conflicts of Interest Statement

The authors report no conflict of interest. Thereis no financial support to report. All authors certifythat all my affiliations with or financial involvementin, within the past 5 years and foreseeable future, anyorganization or entity with a financial interest in orfinancial conflict with the subject matter or materialsdiscussed in the manuscript are completely disclosed(e.g., employment, consultancies, honoraria, stockownership or options, expert testimony, grants or patentsreceived or pending, royalties).