Water, Water Everywhere

Contributed by readers. Edited by Sheldon Jacobson, MD

A 68-year-old woman was brought to the emergency department by her family for treatment of abdominal pain, constipation, and abdominal distension. These symptoms had progressed in severity during the previous three weeks. Because the patient did not speak English, the history was obtained primarily through an interpreter and from her family. She had type 2 diabetes and multiple cardiovascular conditions, including an ischemic cardiomyopathy that developed after several documented myocardial infarctions. She had also had a left-sided cerebrovascular accident one year earlier.

The patient was taking furosemide, and when she sought evaluation of her symptoms from her primary physician several days before being evaluated in the emergency department, he determined that the problem was "her heart" and increased her dosage of furosemide and recommended a low-salt diet. Her other medications included digoxin, atorvastatin, glyburide, metformin, and warfarin.

The patient, who had chronic constipation, had not had a bowel movement for five days. She was also nauseated but did not report any vomiting, recent fever, or chest pain. She was shorter of breath than usual, and her activity was curtailed to the extent that she was only able to go from bed to chair.

On physical examination, the patient was in moderate respiratory distress. She appeared chronically ill but not overly debilitated. Her pulse was 100 and regular; blood pressure, 100/54 mm Hg; respiratory rate, 22; and temperature, 98°F. Pulse oximetry on room air showed 93% oxygen saturation. Her skin was slightly pale and anicteric. Pulmonary examination showed basilar crackles and decreased breath sounds.

Cardiac examination revealed a regular rhythm, grade 3/6 apical blowing systolic murmur, and an S3 gallop. Her distended abdomen displayed mild, diffuse tenderness on palpation. The liver was enlarged and tender on palpation. Rectal examination revealed hard heme-negative stool in the ampulla. Massive edema of the legs extending to the upper thighs was present.

An electrocardiogram (ECG) revealed extensive anterior wall injury but was unchanged from her most recent reading. Findings on a chest film were indicative of congestive heart failure (CHF) with small bilateral pleural effusions. Her hemoglobin was 11.5 gm/dL and her white blood cell count was 8,600/mm3. Laboratory tests revealed the following: sodium, 134 mg/dL; potassium, 3.2 mg/dL; chloride, 95 mg/dL; CO2, 22 mEq/L; blood urea nitrogen, 44 mg/dL; creatinine, 2.4 mg/dL; and glucose, 246 mg/dL.

The examining physician was certain that he was dealing with worsening CHF and constipation, and communicated his findings and impressions to the patient's primary physician. As an added precaution to rule out silent ongoing ischemia, a random blood sample was obtained for measurement of cardiac enzyme levels. The creatine phosphokinase (CPK) MB and troponin I levels were normal.

The emergency physician and primary physician agreed that furosemide should be administered intravenously, and they concurred that adding lactulose, metolazone, and spironolactone to the patient's regimen would be a useful, although temporizing, measure. Both physicians agreed that the patient's prognosis was poor and that hospitalization was unlikely to provide additional benefit. The patient was discharged and referred to a home care service.

Three days later the patient grew progressively more distended and more dyspneic and returned to the emergency department, where she was examined by another physician. He proceeded to repeat all of the tests conducted at the last visit, which yielded essentially the same results. In addition, because the abdominal distension was pronounced, an obstruction series was ordered. The films showed a substantial amount of stool in the colon. There was no obvious small bowel obstruction, but a major finding was a ground-glass appearance to her abdomen, consistent with a diagnosis of ascites, with the bowel loops displaced upwardly.

The patient was admitted and received treatment for CHF, constipation, electrolyte imbalances, and ascites. During the second week of hospitalization, an abdominal computed tomography (CT) scan with contrast was obtained, disclosing a huge pelvic and lower abdominal mass. On the third week of hospitalization a laparotomy was performed, which revealed extensive ascites and a huge left adnexal cystic mass. Because the patient had postoperative pulmonary complications, she required prolonged intubation, but her condition improved significantly. After three months of evaluation through a rehabilitation service, the patient gradually was able to return to the quality of life she had enjoyed before the onset of the presenting symptoms. Pathologic evaluation of the mass identified it as a benign cystadenoma of the ovary.

COMMENT

This case highlights the danger of making assumptions about the condition of elderly patients who have several concurrent disorders. Many physicians tend to assume that there is little to be done for a frail patient with a terminal disorder except keep him or her comfortable, rather than look for additional disease that may be causing the symptoms. The danger of this assumption is compounded when the original diagnosis is not reconsidered after the patient's condition worsens. A decline in the patient's condition only serves to confirm the assumption that the disease has progressed; other diagnostic possibilities often are not entertained.

Although the initial emergency physician in this case did attempt to rule out recent myocardial infarction, he did not pursue alternative diagnoses when the cardiac enzyme levels turned out to be normal. It is true that progressive CHF can be caused by worsening of the underlying heart disease, but one must also consider other, usually treatable disorders that can aggravate or mimic CHF.

The physicians caring for this patient used the laws of parsimony to limit their focus to the existing disorders as the total explanation for all of her problems. However, elderly patients in particular may have simultaneously several discreet, unrelated, and life-threatening conditions, and the differential diagnosis should be no less thorough in patients with preexisting disease, particularly when their condition has recently changed significantly.

The first issue is to decide whether there is a cardiac basis for the anasarca. Thus, one should check for silent mitral and aortic stenosis, endocarditis, papillary muscle rupture, myocarditis, cardiac tamponade, constrictive pericarditis, and any treatable cause of cardiomyopathy. Cardiac tamponade could be caused by pericarditis, coagulopathy, or left ventricular rupture resulting from a transmural infarct or myocardial aneurysm. Potentially reversible causes of cardiomyopathy include myxedema, sarcoid, hemochromatosis, cardiotoxic effects of medications, thyrotoxicosis, acromegaly, and excessive ethanol ingestion.

Factors that could adversely affect the condition of a patient with CHF whose disease was previously stable include noncompliance with medical therapy, increased salt intake, worsening disease, and progressive renal insufficiency. Other possibilities include anemia and high-output failure, such as would occur with arteriovenous malformations or fistulae or with thiamine deficiency and hyperthyroidism. Electrolyte disturbances, such as hypocalcemia and hypophosphatemia, also could worsen CHF. Worsening right-sided failure caused by cor pulmonale secondary to chronic obstructive pulmonary disease or pulmonary embolism should be considered as well. Cardiac tamponade and constrictive pericarditis would also present as primarily right-sided heart failure.

Noncardiopulmonary conditions that mimic CHF include ascites and large intraabdominal masses. These conditions prevent normal diaphragmatic function and thus cause shortness of breath. In addition, the ascites of cirrhosis and nephrosis are usually also associated with generalized edema. Cryptogenic ascites may be caused by unsuspected cirrhosis, hepatic vein thrombosis, myxedema, peritoneal tuberculosis, pseudomyxoma peritonei, and peritoneal carcinomatosis.

Large intraabdominal masses, particularly when they are ovarian in origin, cause symptoms similar to those of cryptogenic ascites. The mass can obstruct venous or lymphatic return from the lower extremities and may be associated with peripheral edema. Differentiation between abdominal masses and ascites can best be accomplished via abdominal ultrasound examination, and the mass further delineated by means of CT scanning.

Fainting Spells

A 29-year-old man was brought to the emergency department by ambulance after he fell approximately 15 feet from a scaffold. The patient, a mason, was pointing a stone wall when he became dizzy and slipped off the platform. A coworker who had witnessed the incident noted that the patient landed on his back and right elbow. Although the patient initially appeared somewhat "dazed" at the scene, he was alert and answered questions appropriately. The right elbow showed obvious injury. The man reported having pain and tenderness over his entire back and severe pain in the elbow.

When the ambulance arrived, paramedics recorded a blood pressure of 130/90 mm Hg; pulse, 100; and respiratory rate, 20. They bandaged and splinted the right elbow and immobilized the patient on a long board.

The patient's medical history was unremarkable. He did not report using recreational drugs or alcohol and was not taking any medications. He noted that he and several family members were "fainters," however.

On evaluation in the emergency department, the patient's airway was stable. Cervical tenderness was not present, and results of the screening cardiopulmonary examination were normal. Trauma examination revealed moderate tenderness along the right thoracic and lumbar paraspinal areas and a deep laceration over the right olecranon. Range of motion was normal in the right hand and wrist but markedly reduced in the right elbow, which was held fixed at 90º of flexion. Sensation was reduced over the ulnar distribution of the right-hand capillary refill.

The examining physician determined that the patient had sustained only a possible fracture of the elbow and that his other injuries were minor. The patient was demobilized and plain films of his elbow and chest were obtained. The images revealed only a comminuted fracture of the olecranon. The on-call orthopedist was summoned and the patient was prepared for surgery.

While being helped to the bathroom before being taken to surgery, however, the patient became unresponsive and had a very brief, generalized tonic-clonic seizure. During the seizure, he fell, striking his head on the asphalt floor tile.

The patient was placed in a supine position. Vital signs were measured again, including a blood pressure of 80/66 mm Hg and a pulse oximetry reading of 96% saturation on room air. A fluid bolus was given, and the patient was placed in the trauma resuscitation bay. He gradually regained consciousness and, once alert, reported having a headache. After administration of a liter of saline over a 20-minute period, his blood pressure was 110/60 and his pulse was 56 and regular. Other than a contusion over the occiput, no new changes were found.

The attending physician initiated an official trauma call, summoning a surgical chief resident and his team to the trauma room. The trauma team noted that the patient had sustained head trauma and was hypotensive. On the basis of these findings, and to rule out internal bleeding before the elbow could be pinned, a CT scan was performed. The findings on CT were normal, however. The patient was then examined by a neurosurgeon and a neurologist. Although this examination further delayed surgical repair of the fracture, it did not reveal any significant findings.

In all, because the emergency department was very busy, the testing and consultation took almost six and a half hours, and the emergency physician believed he was powerless to speed things up. By the time the patient was cleared for surgical repair of the open fracture, the long and unnecessary delay had significantly increased the potential for infection. Fortunately, the postsurgical recovery was uneventful.

COMMENT

An analysis of this case reveals much to consider. Viewed in hindsight, the cascade of events are as follows: the patient sustained an open fracture to his elbow and had vasovagal syncope while assuming an upright position. The syncope was interpreted as a hypotensive episode caused by possible internal bleeding; the seizure was thought to be a sign of underlying traumatic brain injury. But the patient had already revealed that he was a fainter. That pertinent information and the association of bradycardia with hypotension should have led to the diagnosis of benign vasovagal syncope. However, because the attending physician on duty had never seen brief anoxic seizures associated with syncope, he assumed the worst.

Actually, brief seizures with syncope are not unusual, particularly if a patient cannot assume the supine position quickly. The patient in this case was being supported in the erect position. He never had orthostatic testing of his vital signs because he remained relatively hypotensive for approximately 20 minutes, at which point the staff was too wary of his previous fainting spell to remove him from bed to perform the necessary tests.

The other issue in this case was the loss of control of the care process by the attending physician on duty. The surgical team performed a full textbook evaluation. The emergency physician was insecure and therefore did not call a halt to the proceedings, and the trauma team was focused on completing the evaluation. The test of a true team leader is not only knowing when to start a major trauma evaluation but also knowing when to call a halt and move on to the main business at hand.

Making the decision to halt the full examination, however, is a difficult one. If the physician is wrong and there is another cause of the symptoms that is overlooked as a result, the physician could face investigation for failure to adequately evaluate the patient. In retrospect, calling the trauma team was not productive--although this patient's mechanism of injury was potentially serious. In cases such as this in which there are a lot of possibilities, clinical judgment has to supervene and dictate the extent of the evaluation.