In July 1995, an outbreak of adenovirus infection occurred in a chronic psychiatric care facility in Rhode Island. At the time of the outbreak, there were 53 residents and more than 200 full- and part-time staff members. All of the residents and 33 of the staff were screened for infection. Adenovirus infection, defined as a 4-fold or greater change in antibody titer and/or isolation of adenovirus from the respiratory tract, was found in 24. Eighteen people (14 residents, four staff) developed radiologic evidence of pneumonia. Of these 18, 17 (94%) had adenovirus infection, and 14 (13 residents, one staff) required hospitalization. Seven patients were admitted to our institution, six were treated at the acute care ward of the chronic health care facility, and one was admitted to another area hospital.

The first of the seven patients admitted to our hospital presented before the outbreak had been discovered. He was treated on the general medical ward and discharged. The next three patients were brought to the emergency room at the same time and admitted to the ICU. Their simultaneous presentation led us to suspect a contagious disease, and it initiated the investigation of the outbreak. Two more patients were admitted to the ICU from the emergency room over the next week. The last patient presented 10 d later. Her symptoms were less severe and she stayed in the ICU for only a day. The six patients admitted to our ICU form the basis of this report. The seven patients admitted to the other health care facilities had less severe disease and did not require intensive care.

The medical histories and clinical presentations of each patient are summarized in Table 1. Five patients were residents of the mental health center and were being treated for chronic psychiatric disorders. The other patient was a nurse at the facility. The residents lived in the same building, but they were not roommates. They were ill for 1 to 7 d prior to admission and were treated with antibiotics and oxygen. They were transferred to our hospital because of worsening chest radiographs and increasing oxygen requirements. At the time of admission, most patients were febrile, tachypneic, and tachycardic, but all had normal blood pressure. Chief complaints were dry cough and increasing dyspnea. Chest auscultation revealed rhonchi, rales, or wheezes. Findings suggested unilateral consolidation in one patient. No patient had pharyngitis, conjunctivitis, or hepatomegaly, but one patient had an erythematous, papular rash on her forearms and abdomen. Mental status was not altered from baseline in any of the patients. Initial chest films showed patchy or confluent unilateral air-space opacities, usually in the lower lobes. One patient had a large left pleural effusion (Figure 1).

Fig. 1. Portable upright chest film of Patient 1 breathing spontaneously on the day of admission. Dense consolidations of left lower lobe and lingula are present as well as a left-sided pleural effusion. The right lung and left upper lobe are spared.

Download Figure | Download Powerpoint

[More] [Minimize]

Laboratory abnormalities are shown in Table 2. Despite a normal or decreased white blood cell (WBC) count at the time of admission, most patients had toxic granules, Dohle bodies, vacoulated neutrophils, and an increased percentage of band forms. All of the patients had a drop in WBCs within days of being admitted, but no patient became neutropenic. Except for the patient with chronic renal failure, hematocrit, platelet count, and coagulation times were near normal on admission. Two patients had elevated d-dimers, but fibrinogen levels were normal. Serum transaminase, LDH, and CPK levels peaked within the fourth hospital day, whereas liver enzymes tended to peak later. Fractionation of CPK revealed 100% MM in all cases.

The first three patients were suspected of having Legionella pneumophila and were quarantined in the same room. All but the last patient were treated with ceftriaxone and high doses of erythromycin. Rifampin was added for additional Legionella coverage, and doxycycline was added to cover other atypical pathogens. Most patients remained febrile for at least 48 h. Rectal temperatures rose to 103 to 104° F daily and had little response to acetaminophen. Cooling blankets and ibuprofen were more effective.

All of the patients required supplemental oxygen when they arrived in the ER. Hypercapnea was present in two patients with underlying lung disease: one had chronic bronchitis and the other had a partial pneumonectomy. Respiratory failure developed rapidly in five patients. Ventilator requirements and arterial blood gas determinations are shown in Table 3. Four patients developed diffuse pulmonary infiltrates and met criteria for the Acute Respiratory Distress Syndrome (ARDS) as defined by the American-European consensus conference on ARDS (5). They required high Fi _O2 , positive end-expiratory pressure (PEEP), and paralysis to maintain adequate oxygenation.

Septic shock developed after the onset of respiratory failure in each patient with ARDS. Pulmonary artery catheterizations revealed markedly elevated cardiac outputs and severely decreased systemic vascular resistances (Table 4). An echocardiogram, done on one patient to exclude cardiac tamponade, revealed a hyperdynamic left ventricle, normal valvular function, and no pericardial effusion. Serum bicarbonate fell in three patients, but it was out of the normal range in only one. Renal insufficiency developed in one patient aside from the patient with chronic renal failure.

Diagnostic evaluations included routine Gram stain and bacterial cultures of blood and urine. Urine was tested for Legionella pneumophila antigen and serum samples were sent for cold agglutinins and acute and convalescent antibody titers against Chlamydia psittaci, Coxiella burnetii, and adeno-, echo-, coxsackie, and respiratory syncytial viruses. Sputum was stained and cultured for bacteria, mycobacteria, and fungi. In addition, direct fluorescent antibody (DFA) staining was done for Legionella and Pneumocystis carinii, and cultures were sent for Legionella, Mycoplasma, and respiratory viruses. Two patients underwent fiberoptic bronchoscopy. Diffuse endobronchial erythema and thin mucoid secretions were noted. Bronchoalveolar lavage (BAL) specimens were obtained from both patients. One patient also had an open lung biopsy. The BAL specimens, lung biopsy, and pleural fluid were stained and cultured as described above for sputum.

Adenovirus was isolated in five of the six patients, three in sputum alone, one in BAL fluid, and one in BAL fluid and lung tissue. The one patient without positive cultures for adenovirus had a 6-fold increase in adenovirus antibody titers. Serology for adenovirus infection rose at least 4-fold in four of the five patients in whom adenovirus was isolated. All other smears and cultures were negative except for one patient in whom Mycobacterium tuberculosis was isolated along with adenovirus from a BAL specimen. The smear from the BAL was negative for acid-fast bacilli (AFB), and M. tuberculosis was not isolated from sputum samples. This patient also had a lung biopsy specimen that was culture-positive for adenovirus, but not for M. tuberculosis. Histologic examination of the lung biopsy showed smudge cells consistent with viral pneumonia, but no virus particles were identified by electron microscopy. All positive culture specimens from these patients were identified as serotype 35 by the Centers for Disease Control and Prevention.

Five of the six patients recovered and were discharged. The patient with chronic renal failure prior to admission died after 2 mo in the hospital while still intubated. Despite the highly contagious nature of AVP, there were no known cases of transmission of adenovirus infection to any of the health care providers at Rhode Island Hospital.