Flu: Georgia on CDC's mind

A day or two after CDC's Morbidity and Mortality Weekly Reports (MMWR) released a report about risks to pregnant women from pandemic 2009 flu, CDC held a suddenly announced press briefing about the current H1N1 situation (I listened in but a transcript should be up on the site by the time you read this; check this page). The occasion for the briefing was a worrisome increase in hospitalizations and deaths in CDC's Georgia backyard. Despite housing CDC, Georgia has one of the lower flu vaccination rates in the country and now is experiencing an unexpected recrudescence of H1N1 flu, with numbers of hospitalizations not seen in the state since the height of the pandemic last October. The cases were described as "adults," many with pre-existing medical conditions, with a geographic distribution that, on a preliminary view, might be hitting areas less hard hit than the fall wave of cases. Since late February, Georgia was seeing more hospitalizations (in numbers) than any other state in the union and the reason isn't known. What we know so far is that it is the 2009 H1N1 pandemic, laboratory confirmed and said not to be different than before. What that is based on we don't know. It is not either of the previous seasonal flu strains, seasonal H1N1 or seasonal H3N2, nor influenza B, which is circulating at many locations at very low levels.

Is this the harbinger of a third wave that will spread more generally? It is uncharacteristically late in the usual flu season for this to happen, although this particular flu did poke its head above water just about this time of year last year. Maybe there is something about the virus that makes its seasonal pattern different. Perhaps it's because Georgia citizens have less herd immunity than elsewhere, a not too thinly veiled suggestion by CDC's Dr. Anne Schuchat on the conference call. She did use the opportunity to urge reporters to remind their readers that a flu vaccine, which she noted has an excellent safety record, is the best protection against influenza. This is a CDC Talking Point that has the added virtue of being true. At one point she suggested many of the seriously ill had not been vaccinated, although she gave not data. Back to the question, What does this mean, the truthful answer is, "We don't know."

Asked why the estimated 12,000 deaths for pandemic 2009 H1N1 didn't indicate that this was a much milder flu season than the usual estimate of 36,000 flu deaths in other years, Schuchat made the observation that 11,000 of the 12,000 deaths were in people under the age of 65, a much higher number (not just proportion) than usual. This is another way of saying that the difference is in the epidemiology, i.e., the pattern of disease and death in the population. It's also true, however, that the 36,000 death figure is not just an average, but an estimate of excess mortality calculated in a particular way and that it represents the estimated excess averaged over 30 seasons. Calculated that way, in some seasons the excess is essentially zero, even though in those years many people's deaths would not have happened had they not had the flu (if you want to know more, here's one of a number of posts we wrote on the subject).

And it's true, too, that whatever the numbers, some people are at much more risk than others. Like pregnant women with no other risk factors. The MMWR Report has a vignette of one tragic case. It could have been titled, Why you should get your flu shot:

A woman aged 27 years who was at 32 weeks' gestation (Table) went to her primary care physician during May 2009 after 1 day of fever and cough (Figure). She was treated with antibiotics for 3 days without improvement. Five days after symptom onset, she went to the emergency department, reporting persistent fevers, chills, cough, wheezing, and an episode of near-syncope. On admission she was afebrile, with a respiratory rate of 22 breaths per minute, a heart rate of 96 beats per minute, blood pressure of 100/70 mmHg, and oxygen saturation of 99% on room air. A chest radiograph revealed bilateral lobar pneumonia, and she was treated for community-acquired pneumonia. On hospital day 2, she developed fever to 102.9°F (39.4°C), tachycardia (141 beats per minute) and severe respiratory distress. ARDS was diagnosed, and the patient was transferred to the ICU for mechanical ventilation and treated empirically with oseltamivir, 75 mg twice daily. Rapid influenza diagnostic tests performed on nasopharyngeal specimens 1 day before hospital admission and on hospital day 3 were negative for influenza.

On hospital day 4, because her oxygen saturations worsened to approximately 75% despite maximal ventilation settings, an emergency cesarean delivery was performed. During the procedure, the patient was hypotensive and required multiple blood transfusions. Cultures from bronchoalveolar lavage collected the previous day grew Acinetobacter baumanii. On hospital day 11, diagnosis of 2009 H1N1 was confirmed from a nasopharyngeal swab specimen submitted to the DOHMH Public Health Laboratory on hospital day 3. On hospital day 16, because of refractory hypoxemia and severe ARDS, the woman was transferred to another hospital ICU for extracorporeal membrane oxygenation (ECMO), and oseltamivir was increased to 150 mg, twice daily. Her subsequent hospital course was complicated by volume overload, septic shock, and ventilator-associated pneumonia with Klebsiella pneumoniae and A. baumanii. She died on hospital day 38, a total of 42 days after symptom onset (Figure). At birth, her infant weighed 1,500 g and had Apgar scores of 1 at 1 minute and 1 at 5 minutes after birth; the infant stopped breathing, and neonatal resuscitation efforts were unsuccessful. (MMWR, CDC)