Isolate ICU Patients Until H1N1 Test Results Return

SAN DIEGO — Antigen testing for the pandemic influenza A(H1N1) virus using nasopharyngeal swabs is not sensitive enough to be reliable in ICU patients, the experience at one busy inner-city New York hospital suggests.

Just 3 of 15 critically ill adults admitted to the ICU at Lincoln Medical Center in the South Bronx with H1N1 flu during the May-July outbreak had a positive nasopharyngeal swab. The other 12 tested positive only on bronchial washings or tracheal aspirates, Dr. Raghu S. Loganathan reported at the annual meeting of the American College of Chest Physicians.

Lincoln Medical Center has one of busiest emergency departments in the country, serving a disadvantaged minority population particularly hard hit by 2009 H1N1. The hospital's experience during the first wave of H1N1 infection in late spring/early summer led to a change in ICU practice at the hospital—one deserving of consideration elsewhere, said Dr. Loganathan, director of the medical ICU and stroke center there.

It can take 24–48 hours to obtain a lower respiratory tract specimen and diagnose H1N1 influenza in ICU patients. So now patients admitted to the Lincoln ICU with respiratory symptoms are placed in isolation with droplet and contact precautions, at least until a lower respiratory specimen is obtained and the test results are known.

During the May-July outbreak, when this practice was not yet in place, unprotected health care workers and other ICU patients were exposed to the virus during that initial 24–48 hours, and if test results came back positive for H1N1, they required prophylaxis with oseltamivir (Tamiflu).

The 15 patients with H1N1 flu admitted to the ICU averaged 49 years of age, and were among 40 adults hospitalized with the infection during May-June, for a 37.5% ICU admission rate.

Of the 15 patients, 13 had at least one comorbid condition, including overweight or obesity in 9, diabetes in 8, and asthma or chronic obstructive pulmonary disease (COPD) in 7. These comorbidity rates are substantially higher than those cited for ICU patients with H1N1 flu in other recent series from Mexico, Canada, Australia, and elsewhere in the United States, probably reflecting the demographics of the South Bronx, Dr. Loganathan observed.

Four ICU patients had normal chest x-rays. Eight had unilateral radiographic lung abnormalities, in contrast to the typically bilateral abnormalities seen in patients with seasonal flu. The radiographic findings consisted of alveolar consolidation in seven patients, interstitial infiltrates in two, and nodular infiltrates in two.

Lymphopenia was present in 10 of 15 patients, with a median value of 600 cells/mL. Elevated creatine phosphokinase was detected in nine patients, with a median value of 563 U/L. The median lactate dehydrogenase level in the 15 patients was 362 U/L, with 7 patients having a significantly elevated level indicative of hemolysis.

Mechanical ventilation was used in 12 patients, including 7 with acute respiratory distress syndrome; 10 patients had severe sepsis or septic shock.

The mean ICU length of stay was 9.6 days. Four of the 15 patients died.

According to a recent report, the mean ICU length of stay for critically ill patients with seasonal flu was just 3 days (J. Clin. Virol. 2009;46:275-8), compared with 9.6 days for the H1N1 flu patients, and the in-hospital mortality rate was 18.8% for the seasonal flu group, compared with 26% for the New Yorkers with severe H1N1 influenza.

Just 3 of 15 critically ill adults with H1N1 influenza had a positive nasopharyngeal swab.

Source DR. LOGANATHAN