html xmlns=”http://www.w3.org/1999/xhtml” xml:lang=”en”>
Federal Initiatives in Emergency Preparedness
James M. Blum, MD
Objectives
- Understand the need for critical care data collection in public health emergencies using the 2009 H1N1 swine-lineage as an example
- Ascertain the methods and limitations of existing methods of data collection in critically ill patients
- Describe the high level plans for real-time reporting of critically ill patients in the event of a public health emergency
Key words: public health emergency, critical care reporting, influenza, pandemic
Example of the Problem
On April 15, 2009, the first case of 2009 swine-lineage H1N1 (2009 H1N1) influenza was identified in a 10-year-old child in California as part of a clinical study. Two days later, the second case was identified 130 miles away in an 8-year-old child.1 On April 18, this virus was reported to the World Health Organization (WHO) according to international health regulations. This new strain of influenza was a menagerie of genes from North American and Eurasian swine-lineage H1N1.2
While it had been decades since a worldwide influenza pandemic, 2009 H1N1 was not the virus that was expected to be the culprit. Avian influenza A (H5N1) was a known entity in poultry in certain parts of the world and occasionally infected people, frequently resulting in multiple-organ dysfunction syndrome and death.3 Previous governmental planning surrounding a pandemic was, hence, designed around a very different threat than what was at hand. Little was known about how 2009 H1N1 would affect the human population.
Although 2009 H1N1 was a new entity, many of the mechanisms to address the pandemic functioned well. By April 22, the virus had already been isolated for vaccine development, the US Centers for Disease Control and Prevention (CDC) publication Morbidity and Mortality Weekly Report (MMWR) issued notice of the virus and requested that all influenza that could not be subtyped be sent to the CDC for analysis, and an emergency operations center operated by CDC was opened.
These events occurred just in time for the spread of the virus: The next day 2 samples from Texas were determined to be 2009 H1N1 along with 14 samples from Mexico that were collected prior to April 15.4 By April 25, cases had been reported from New York, Kansas, and Ohio, and the WHO director general declared the outbreak a public health emergency of international concern. On April 26, 11 million doses of antiretrovirals were released from government stockpiles for treatment of patients with suspected H1N1 infection. This came with emergency use authorizations for the medication.5
There was little knowledge of what effect these medications would have on the natural history of the infection. It was known that the virus was susceptible to the medications from in vitro assays, but historically, many physicians did not prescribe antiretroviral medications to healthy patients for seasonal flu given the minimal impact of such medications on the course of the disease.6 Now it was unclear who should receive this medication, although the emergency use authorization suggested that patients who were hospitalized with H1N1 or who had been symptomatic for 2 days would be good candidates. The release of the medications was not accompanied by any method of data collection on the physiological status of patients infected with H1N1.
On April 30, 2 weeks into the outbreak, MMWR reported the initial outbreak in Mexico and described what appeared to be an outbreak in a New York City high school.7 The extent of what was documented regarding the patients’ status was that they presented with respiratory symptoms, fever, and diarrhea.
It was not until May 8 that it started to become clear who was most susceptible to 2009 H1N1.8 In Mexico, almost 12,000 suspected patients had been identified, 949 had been confirmed by laboratory analysis, and 42 had died. In the United States, a total of 1,487 confirmed and probable cases were identified. It was disclosed that 57% of reported cases were in patients aged 5 to 24, and 41% of hospitalized patients were older children or young adults. That this age group would be the at-risk population had not been anticipated from prior flu pandemic planning.
On May 12, nearly a month into the pandemic, the first glimpse of what the critically ill patient with H1N1 in the United States looked like appeared in an MMWR dispatch describing a 33-year-old pregnant woman with confirmed 2009 H1N1 who developed acute respiratory distress syndrome (ARDS) and subsequently died.9 The dispatch did not describe additional characteristics of her illness. The dispatch discussed how pregnant women tended to be at increased risk for influenza, and it encouraged the use of oseltamivir to treat this population.
Six days later, an MMWR early release discussed data on 30 hospitalized patients with known or suspected H1N1, 6 of whom required critical care.10 Four patients required mechanical ventilation, and 3 of the patients required “prolonged” critical care. However, additional information about the patients’ status and the critical care services delivered was provided only for selected patients in case report form.
Finally, on July 10, 2009, an MMWR dispatch was released regarding intensive care patients at the University of Michigan who had laboratory-confirmed 2009 H1N1.11 At this point, 33,902 cases in the United States had been reported, 170 of which were fatal. The University of Michigan report described 10 patients from a tertiary care ICU and extracorporeal membrane oxygenation (ECMO) referral center who were transferred to the center between May 26 and June 18. The report documented the age of the patients as 21 to 51 years; 9 of 10 patients were obese, 7 were morbidly obese, and 5 had pulmonary emboli. Nine of these patients had multiple-organ dysfunction syndrome, and 3 died. Of the 3 who died, time from illness onset to death was between 17 and 30 days, and autopsies on 2 patients confirmed bilateral hemorrhagic pneumonitis, diffuse alveolar damage, and bilateral pulmonary emboli.
The report also discussed the laboratory data (elevated white blood cell count, elevated aspartate aminotransferase, elevated creatine kinase), hypercoagulable complications, and the nature of the therapy the patients received, including the use of vasopressors and advanced forms of ventilation (high-frequency oscillatory or bilevel ventilation with mean airway pressures between 32 and 55 cm H2O). Two patients received venovenous ECMO, and 6 received renal replacement therapy. All 10 patients received oseltamivir at 150 mg twice daily and amantadine beyond 5 days.
Unfortunately, almost 3 months into the pandemic, this was the extent of the information that was available regarding the US critical care experience for H1N1 in 2009. The editorial note that accompanied the report discussed the potential hypercoagulable state that was reported by the authors and provided the observation that 2009 H1N1 patients with ARDS were at very high risk for pulmonary emboli. This was extrapolated by many into the routine use of aggressive anticoagulation for critically ill H1N1 patients at the time. The editorial comment also stated,