The Emergency Department Observation Unit (EDOU) provides a viable alternative to inpatient admission for the management of many acute gastrointestinal conditions with additional opportunities of reducing resource utilization and reducing radiation exposure. Using available evidence-based criteria to determine appropriate patient selection, evaluation, and treatment provides higher-quality medical care and improved patient satisfaction. Discussions of factors involved in creating an EDOU capable of caring for acute gastrointestinal conditions and clinical protocol examples of acute appendicitis, gastrointestinal hemorrhage, and acute pancreatitis provide a framework from which a successful EDOU can be built.
Key points
- •
The Emergency Department Observation Unit (EDOU) provides a viable alternative to inpatient admission for the management of many acute gastrointestinal conditions with additional opportunities of reducing resource utilization and reducing radiation exposure.
- •
Using available evidence-based criteria to determine appropriate patient selection, evaluation, and treatment provide higher-quality medical care and improved patient satisfaction.
- •
Descriptions of factors involved in creating an EDOU capable of caring for acute gastrointestinal conditions and clinical protocol examples of acute appendicitis, gastrointestinal hemorrhage, and acute pancreatitis provide a framework from which a successful EDOU can be built.
Mr Smith is a 43-year-old man who presents to the emergency department with a complaint of “gnawing” epigastric pain associated with nonbloody, nonbilious vomiting. He admits to excessive alcohol intake the week prior. The current symptoms are similar to his 2 prior episodes of pancreatitis. Vital signs are normal except for mild tachycardia at 108 beats per minute. On examination, he is in obvious discomfort. He has tenderness in the epigastric region without rebound or guarding. Laboratory studies are significant only for a lipase level of 2200 U/L (normal <300). He is placed in the observation unit where he is treated with intravenous hydration and parenteral opioids. A right upper quadrant ultrasound is unremarkable. The following day his diet is advanced and he is transitioned to oral analgesics. Alcohol counseling is provided and he is discharged home to follow-up at his primary care provider 2 days later.
Introduction
Gastrointestinal (GI) emergencies accounted for nearly 15 million of the 122 million emergency department (ED) visits in the United States in 2007 at a cost of $27.9 billion. Traditionally, the care of a patient with abdominal pain was to perform a rapid diagnostic workup with efforts aimed at relief of symptoms. After the initial ED workup, patients would be either discharged home or admitted to a surgical or medical inpatient service. The ED provider, under pressure to rapidly and accurately diagnose and treat patients, has increasingly relied on early radiographic testing. The difference in imaging utilization for abdominal complaints rose from 19.9% in 1999 to 44.3% in 2008. Some patients may benefit from a period of time during which they may undergo further diagnostic workup, risk stratification, and treatment before a final disposition. ED Observation Units (EDOUs) have the potential to improve diagnostic accuracy with more judicious use of imaging modalities, lower morbidity and mortality, improve the patient experience, and reduce health care expenditures.
The potential for EDOUs to reduce patient exposure to ionizing radiation deserves special emphasis. Recent studies in the trends of computed tomography (CT) use in the ED have revealed a tradeoff between the ability to diagnose high-risk conditions more rapidly at the expense of increased cost and increased exposure to radiation and iodinated contrast agents. This is further compounded by the discovery of incidental findings that necessitate potentially needless additional diagnostic testing. The incidence of adrenal “incidentalomas,” a term coined to denote the presence of an adrenal mass in an otherwise asymptomatic patient, found on CT scans approaches 9%. The risk of ionizing radiation is dependent on the dose, tissue exposed, and age of the patient. A CT scan of the abdomen and pelvis exposes the patient to 10 mSv or 3 times the expected annual natural background radiation dose. This exposure, especially when occurring earlier in life, has been associated with a small increase in interval development of cancer. This risk is small to the individual, but given the magnitude of the population that is exposed, the cumulative impact on increased cancer rates is significant.
EDOUs balance the need to accurately and safely diagnose and treat patients with the pressure to do so rapidly but with judicious use of imaging. This balance is achieved by combining the ED assessment with standardized risk assessment tools and serial examinations, a core principle of observation medicine. For lower-risk cases for which no imaging has been performed as part of the initial ED evaluation, the period of observation provides the opportunity to perform repeat assessments of the patient’s condition and to obtain imaging in the event of clinical worsening or failure to improve.
EDOU protocols pertaining to GI complaints will share many common themes. These include standard inclusion and exclusion criteria that exist to select for the most appropriate patient who would benefit from the time in the unit. These criteria must be individualized to local system needs and updated as systems change or new knowledge in the care of conditions evolves.
Inclusions and exclusions are considered on 3 levels: the health care system level, the general patient level, and within the specific patient protocol. Considerations at the level of the health care system include capacity constraints, availability of required resources, unit factors, and team factors. Risk tolerance is proportional to the ability of a system to provide the resources needed for a change in patient status. An example of a unit factor to consider is the presence of individual rooms versus a communal space that is set apart by curtains. The latter would not be conducive for the care of patients with communicable diseases, diarrhea, or those being prepared for a colonoscopy. The team treating the patients placed in the EDOU should have the required skills and expertise to care for the conditions that are likely to be encountered.
Abdominal pain has an exhaustive differential diagnosis that ranges from very low-risk conditions, such as constipation, to catastrophic conditions, such as a ruptured abdominal aortic aneurysm. General exclusion criteria include patients with unstable vital signs, presence of a surgical abdomen, or severe sepsis, or if the diagnostic workup reveals any emergent surgical condition. Patients who are immunocompromised are often not ideal candidates for placement in the EDOU. The severity of their condition is often not evident on initial evaluation and their care involves utilization of multiple resources. Patients with higher-risk immunosuppressive states, such as those who have received transplantation or those with malignancy undergoing chemotherapy, those with human immunodeficiency virus/AIDS, and those with rheumatologic conditions requiring immunosuppressive agents may have very atypical presentations of GI illnesses, such as peritonitis, opportunistic infections, graft-versus-host-disease, and neutropenic enterocolitis. There are some disease-specific conditions that are not likely appropriate for the EDOU. These include conditions in which it is known that the patient will require a prolonged in-hospital stay, such as severe gastroparesis. High-risk conditions should be thoroughly considered in patients with risk factors before placement in the EDOU. One example is consideration of mesenteric ischemia in patients with a history of atrial fibrillation, cardiomyopathy, and other low-flow states.
The provider should consider that several emergent conditions may initially appear to have a GI etiology. Misleading complaints include epigastric pain of an inferior wall acute myocardial infarction, right upper quadrant pain of hepatic congestion in congestive heart failure, generalized abdominal pain with nausea and vomiting in diabetic ketoacidosis, and back and flank pain in pulmonary embolism.
Abdominal pain is the most common chief complaint among the elderly presenting to the ED. Patients older than 80 have nearly double the mortality of younger patients if the diagnosis is delayed. Due to physiologic changes caused by lax abdominal wall musculature and increased use of medications, elderly individuals often have atypical or muted signs and symptoms, including lack of vital sign alteration, such as fever or tachycardia. Only 17% of elderly individuals with a perforated appendix have classic complaints. Age-related physiologic changes may warrant exclusion of elderly individuals from some EDOU GI protocols.
Behavioral and social factors may preclude a reasonable safe disposition within the usual timeframe for EDOU care. For example, it may be unreasonable to perform an observation protocol for acute pancreatitis on a patient who also has symptoms of alcohol withdrawal due to ongoing behavior conditions that had previously occurred. This may affect the care of the patient, as well as the care of other patients in the EDOU.
Each health care system and ED team must consider a general approach to the patient with GI disorders that reflects the abilities of the system and all of its resources, including consultant availability and provider abilities and resources. General patient considerations of catastrophic conditions, immunosuppression, mimic condition considerations, elderly, and behavioral and social factors must be considered before consideration to place in an observation protocol.
Specific patient protocols use best clinical practices and available evidence to standardize care. EDOUs may be used for the evaluation and possible treatment of appendicitis, GI bleeding, and pancreatitis.
Introduction
Gastrointestinal (GI) emergencies accounted for nearly 15 million of the 122 million emergency department (ED) visits in the United States in 2007 at a cost of $27.9 billion. Traditionally, the care of a patient with abdominal pain was to perform a rapid diagnostic workup with efforts aimed at relief of symptoms. After the initial ED workup, patients would be either discharged home or admitted to a surgical or medical inpatient service. The ED provider, under pressure to rapidly and accurately diagnose and treat patients, has increasingly relied on early radiographic testing. The difference in imaging utilization for abdominal complaints rose from 19.9% in 1999 to 44.3% in 2008. Some patients may benefit from a period of time during which they may undergo further diagnostic workup, risk stratification, and treatment before a final disposition. ED Observation Units (EDOUs) have the potential to improve diagnostic accuracy with more judicious use of imaging modalities, lower morbidity and mortality, improve the patient experience, and reduce health care expenditures.
The potential for EDOUs to reduce patient exposure to ionizing radiation deserves special emphasis. Recent studies in the trends of computed tomography (CT) use in the ED have revealed a tradeoff between the ability to diagnose high-risk conditions more rapidly at the expense of increased cost and increased exposure to radiation and iodinated contrast agents. This is further compounded by the discovery of incidental findings that necessitate potentially needless additional diagnostic testing. The incidence of adrenal “incidentalomas,” a term coined to denote the presence of an adrenal mass in an otherwise asymptomatic patient, found on CT scans approaches 9%. The risk of ionizing radiation is dependent on the dose, tissue exposed, and age of the patient. A CT scan of the abdomen and pelvis exposes the patient to 10 mSv or 3 times the expected annual natural background radiation dose. This exposure, especially when occurring earlier in life, has been associated with a small increase in interval development of cancer. This risk is small to the individual, but given the magnitude of the population that is exposed, the cumulative impact on increased cancer rates is significant.
EDOUs balance the need to accurately and safely diagnose and treat patients with the pressure to do so rapidly but with judicious use of imaging. This balance is achieved by combining the ED assessment with standardized risk assessment tools and serial examinations, a core principle of observation medicine. For lower-risk cases for which no imaging has been performed as part of the initial ED evaluation, the period of observation provides the opportunity to perform repeat assessments of the patient’s condition and to obtain imaging in the event of clinical worsening or failure to improve.
EDOU protocols pertaining to GI complaints will share many common themes. These include standard inclusion and exclusion criteria that exist to select for the most appropriate patient who would benefit from the time in the unit. These criteria must be individualized to local system needs and updated as systems change or new knowledge in the care of conditions evolves.
Inclusions and exclusions are considered on 3 levels: the health care system level, the general patient level, and within the specific patient protocol. Considerations at the level of the health care system include capacity constraints, availability of required resources, unit factors, and team factors. Risk tolerance is proportional to the ability of a system to provide the resources needed for a change in patient status. An example of a unit factor to consider is the presence of individual rooms versus a communal space that is set apart by curtains. The latter would not be conducive for the care of patients with communicable diseases, diarrhea, or those being prepared for a colonoscopy. The team treating the patients placed in the EDOU should have the required skills and expertise to care for the conditions that are likely to be encountered.
Abdominal pain has an exhaustive differential diagnosis that ranges from very low-risk conditions, such as constipation, to catastrophic conditions, such as a ruptured abdominal aortic aneurysm. General exclusion criteria include patients with unstable vital signs, presence of a surgical abdomen, or severe sepsis, or if the diagnostic workup reveals any emergent surgical condition. Patients who are immunocompromised are often not ideal candidates for placement in the EDOU. The severity of their condition is often not evident on initial evaluation and their care involves utilization of multiple resources. Patients with higher-risk immunosuppressive states, such as those who have received transplantation or those with malignancy undergoing chemotherapy, those with human immunodeficiency virus/AIDS, and those with rheumatologic conditions requiring immunosuppressive agents may have very atypical presentations of GI illnesses, such as peritonitis, opportunistic infections, graft-versus-host-disease, and neutropenic enterocolitis. There are some disease-specific conditions that are not likely appropriate for the EDOU. These include conditions in which it is known that the patient will require a prolonged in-hospital stay, such as severe gastroparesis. High-risk conditions should be thoroughly considered in patients with risk factors before placement in the EDOU. One example is consideration of mesenteric ischemia in patients with a history of atrial fibrillation, cardiomyopathy, and other low-flow states.
The provider should consider that several emergent conditions may initially appear to have a GI etiology. Misleading complaints include epigastric pain of an inferior wall acute myocardial infarction, right upper quadrant pain of hepatic congestion in congestive heart failure, generalized abdominal pain with nausea and vomiting in diabetic ketoacidosis, and back and flank pain in pulmonary embolism.
Abdominal pain is the most common chief complaint among the elderly presenting to the ED. Patients older than 80 have nearly double the mortality of younger patients if the diagnosis is delayed. Due to physiologic changes caused by lax abdominal wall musculature and increased use of medications, elderly individuals often have atypical or muted signs and symptoms, including lack of vital sign alteration, such as fever or tachycardia. Only 17% of elderly individuals with a perforated appendix have classic complaints. Age-related physiologic changes may warrant exclusion of elderly individuals from some EDOU GI protocols.
Behavioral and social factors may preclude a reasonable safe disposition within the usual timeframe for EDOU care. For example, it may be unreasonable to perform an observation protocol for acute pancreatitis on a patient who also has symptoms of alcohol withdrawal due to ongoing behavior conditions that had previously occurred. This may affect the care of the patient, as well as the care of other patients in the EDOU.
Each health care system and ED team must consider a general approach to the patient with GI disorders that reflects the abilities of the system and all of its resources, including consultant availability and provider abilities and resources. General patient considerations of catastrophic conditions, immunosuppression, mimic condition considerations, elderly, and behavioral and social factors must be considered before consideration to place in an observation protocol.
Specific patient protocols use best clinical practices and available evidence to standardize care. EDOUs may be used for the evaluation and possible treatment of appendicitis, GI bleeding, and pancreatitis.
Appendicitis
Introduction/Epidemiology
Suspected appendicitis is a common diagnostic consideration in patients who present to the ED with abdominal pain and is the most common diagnosis of abdominal complaints that requires surgery. The lifetime risk of appendicitis is 8.6% for male individuals and 6.7% for female individuals. The initial ED evaluation includes a thorough history and physical examination. Although laboratory evaluation may be undertaken, it is important to note that no single laboratory value can rule-in or rule-out the diagnosis of appendicitis. A retrospective review in 2015 showed that no single value of white blood cell count, C-reactive protein, or their combination resulted in a positive predictive value of greater than 80% or a negative predictive value of greater than 90%. Thus, the diagnosis is considered based on the history, and physical and laboratory values can be used to support but not definitively rule-in or rule-out this diagnosis.
Risk Scores/Imaging
Scoring systems have been developed to aid in the evaluation of patients with suspected appendicitis. These scores often use components of the history, physical, and laboratory values. The most commonly used is the Alvarado score, also known by its mnemonic, MANTRELS score. This score uses 8 predictive factors ( Table 1 ) that were given a point value of 1 or 2 based on diagnostic weight. Other scoring systems that may be used include the Pediatric Appendicitis Score, Appendicitis Inflammatory Response Score, Raja Isteri Pengiran Anak Saleha Appendicitis score, and the Adult Appendicitis Score. These clinical decision tools can be used to risk stratify patients into low, intermediate, and high risk for appendicitis.
| Points | |
|---|---|
| Symptoms | |
| Nausea and/or vomiting | 1 |
| Anorexia | 1 |
| Migration of pain to the right lower quadrant | 1 |
| Signs | |
| Tenderness in the right lower quadrant | 2 |
| Rebound | 1 |
| Temperature ≥37.3°C (99.1°F) | 1 |
| Laboratory tests | |
| White blood cell count 10.0 × 10 9 /L | 2 |
| Left shift | 1 |
Imaging studies are obtained if the clinician has a high enough concern, especially if the clinical scoring system used places the patient in the moderate to high-risk group. CT imaging of the abdomen and pelvis with contrast is considered the most appropriate imaging modality by the American College of Radiology (ACR) for evaluation of adults with a classic clinical presentation for appendicitis. CT findings of acute appendicitis include an enlarged, thickened (wall thickness >3 mm), fluid-filled appendix with or without an appendicolith with associated periappendiceal inflammatory changes. Other findings that may be associated with appendicitis include inflammatory changes of the adjacent bowel, enlarged mesenteric lymph nodes, and free fluid. A drawback to the use of CT scans is the exposure to ionizing radiation and iodinated contrast agents. Despite these concerns, the utilization of CT imaging has increased dramatically. In one single-center review, the utilization of CT before appendectomy increased from 1% in 1990 to 97.5% in 2007. The classically accepted negative appendectomy rate, defined as the rate of pathologically normal appendices surgically removed in patients suspected of having appendicitis, was historically between 15% and 25%. Recent data suggest that this has now decreased to less than 2%.
Ultrasound is increasingly being used as the primary imaging modality in certain cases. The ACR recommends graded compression ultrasound as the initial imaging test in children younger than 14 and in pregnant patients. This technique uses a high-resolution linear transducer placed at the site of maximal tenderness. Pressure is applied to compress and displace bowel loops to identify the appendix. Normally, the appendix is a blind ending tubular structure that is compressible and may demonstrate peristalsis. In cases of appendicitis, the appendix becomes noncompressible and enlarges. Generally, it is considered abnormally large if the outer diameter is >6 mm although in one study of asymptomatic patients, 3.6% of visualized appendices exceeded this cutoff.
Ultrasound is limited by availability and it is highly operator dependent and has lower sensitivity than CT. Ultrasound, although recommended as the first-line imaging test for pregnant patients, has limited value during the second and third trimesters. Lehnert and colleagues retrospectively evaluated 99 patients in the second and third trimesters who underwent sonographic evaluation for suspected appendicitis. The appendix was visualized in only 3 (3%) of the patients. Ultrasound detected an abnormal appendix in only 2 (27%) of the 7 cases that were surgically confirmed.
Observation Unit Management
The benefit of active observation for pediatric patients with concern for appendicitis was demonstrated by White and colleagues in 1975. They compared 2 groups of consecutive pediatric patients, the first receiving appendectomy and the second receiving a “protective approach” of active observation. The use of active observation reduced the negative appendectomy rate from 15.0% to 1.9% without a significant increase in the rate of perforation. This study was followed up in 1986 by a study by Thomson and Jones. In this single-center study, there were 153 patients not treated with initial surgery who were placed in active observation where they underwent repeated examinations. Eighteen of these patients eventually had surgery, with 7 cases of appendicitis. The remaining 135 patients had no surgical intervention. Although the details reported are limited, the authors state that “no patient suffered as a result of this policy” and they were able to manage 88% of indeterminate cases nonoperatively.
Currently, the primary use of an observation unit for the care of a patient with suspected appendicitis is in cases of diagnostic uncertainty but lower risk using clinical decision tools. Some patients with a low score may benefit from being placed in observation and having serial clinical examinations and laboratory studies performed. This approach conforms to the radiological principle of ALARA (As Low As Reasonably Achievable) in which the providers reduce the exposure to radiation as much as possible. This may be especially beneficial in the pediatric population, which is more radiosensitive than their adult counterparts. Another beneficial use of the observation unit is for patients who undergo imaging as part of their initial diagnostic workup but the results are either negative or indeterminate in a symptomatic patient. Wai and colleagues performed a retrospective analysis of pediatric patients admitted to a single-center’s EDOU. All patients placed in the EDOU had an initial evaluation that included nondiagnostic imaging. Of those, 16% were eventually admitted to the hospital, with 9% undergoing surgery. Of those who underwent surgery, more than half had appendectomies performed. The negative appendectomy rate in this group was 38%. The higher than traditional rate is likely due to preselection for a more diagnostically challenging population.
A retrospective study by Graff and colleagues assessed the probability of appendicitis using the Alvarado score before placement in an observation unit compared with the score at the end of the observation period. Patients determined to have appendicitis had a change in average score from 6.8 to 7.8, whereas patients without appendicitis had a decrease from 3.8 to 1.6. Another study compared the use of an EDOU as compared with a Surgical Assessment Unit (SAU). The time to decision for surgery and time to discharge from the hospital was not different between the EDOU and the SAU. This suggests that ED providers are as capable as surgeons at managing these indeterminate patients.
Recently, there has been debate about using a nonoperative, antibiotics-only approach for the treatment of select cases of appendicitis. The Non Operative Treatment for Acute Appendicitis (NOTA) and Antibiotic Therapy versus Appendectomy for Treatment of Uncomplicated Acute Appendicitis (APPAC) trials were published in 2014 and 2015, respectively. Nonoperative treatment of appendicitis remains controversial and would not currently be considered usual practice in a typical EDOU.
Gastrointestinal bleeding
Introduction
Gastrointestinal bleeding (GIB) is usually categorized as upper gastrointestinal bleeding (UGIB) or lower gastrointestinal bleeding (LGIB), with the distinction made anatomically at the ligament of Treitz, the ligament that suspends the duodenojejunal flexure and inhibits the reflux of gastrointestinal contents distal to this site. Thus, hematemesis usually indicates bleeding from the upper GI tract. GIB can vary in clinical severity, risk of mortality, treatment requirements, length of stay, and cost. Several clinical risk scores have been developed to assist the provider in recognizing and managing these factors. GIB protocols with a standard approach to assessment and treatment occurring in an EDOU setting can improve these factors.
Epidemiology
In the United States, 48 to 160 per 100,000 people have a UGIB annually, whereas 20 per 100,000 have an LGIB. Common causes of UGIB include peptic ulcer disease, erosive disease, variceal bleeding, esophagitis, and Mallory-Weis tear. Common causes of LGIB include diverticular bleeding, hemorrhoids, polyps, colorectal cancer, intestinal ischemia, angiodysplasia, and colitis.
Clinical Presentation
On presentation to the ED, the patient will undergo evaluation to determine the cause and severity of the GIB and urgent issues such as impending shock will be rapidly stabilized. A careful GIB history will include a description of hematemesis, stool color and frequency, symptoms of hepatic disease, and symptoms of blood loss, including syncope. Comorbid conditions must be noted, including previous GIB and endoscopy results, as well as use of nonsteroidal anti-inflammatory drugs (NSAIDs) and anticoagulants, as they are common causes of GIB. Evaluation includes obtaining orthostatic vital signs, examination of any vomitus produced, and assessment of stool for evidence of blood. Signs of liver failure suggest the possibility of a variceal bleed and coagulopathy. Variceal bleeding is high risk and should be excluded from an EDOU GIB protocol. Nasogastric lavage can help identify UGIB when positive, but has not been found to exclude UGIB when negative. Routine laboratory testing includes hemoglobin, blood urea nitrogen (BUN), creatinine (Cr), coagulation profile, and a type and screen. Blood transfusion should not be delayed for laboratory results in suspected impending shock. An elevated lactate greater than 2.5 mmol/L is associated with hypotension within 24 hours.
When the source of GIB is not initially apparent, several factors may help increase the likelihood of correctly diagnosing the source. An upper GI source for bleeding is correlated with a history of passing black stool, presence of melena, history of prior UGIB, positive nasogastric tube aspirate and BUN to Cr ratio greater than 30. Factors reducing the likelihood of UGIB include presence of blood clots in stool, history of LGIB, and BUN to Cr ratio less than 30.
Patients with GI bleeding can be successfully categorized into risk categories using clinical scores, such as the Clinical Rockall Score (CRS) and the Glasgow Blatchford Score (GBS), which can help define the likelihood of mortality, need for blood transfusion, risk for rebleeding and need of urgent endoscopy in UGIB. In 2000, Blatchford and colleagues identified clinical criteria that comprise the GBS, which correctly identified 99% of serious UGIB needing treatment. This study was validated in a multicenter trial showing that all patients sent home with a Blatchford Score of 0 had no need for intervention on follow-up. This validation study also showed a reduction in admission rate for UGIB from 96% to 71%. A comparison of patients with low-risk scores by GBS or CRS showed the rate of rebleeding for both scoring systems to be 13%. A retrospective analysis of patients placed in an EDOU found that only 5% of patients identified as low risk by GBS had a rebleeding episode within 90 days. Thus, GBS and CRS can help stratify patients into low and high risk, but these should be used with caution when considering sending a patient directly home without endoscopy or further observation ( Table 2 ).
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
