From Medscape Infectious Diseases
John G. Bartlett, MD
It may be premature for this analysis since we have had 2 waves of influenza and no one can exclude the possibility of a third wave. Nevertheless, it is a good time to gather the major lessons learned to date. My "top 10" list (in no particular order) is the following:
1. Current surveillance systems do not seem to work. Before the 2009 (H1N1) pandemic began, it was believed that the next pandemic would come from Asia, where surveillance systems were established on the basis of this prediction. The next pandemic was supposed to be a new strain of influenza such as H5N1, and it was expected to be associated with a high mortality rate. Instead, the pandemic came from Mexico, it involved a variant of H1N1 (the oldest of all known strains), and the mortality seemed extraordinarily low, about 0.02% compared with mortality from Spanish flu, which at 2.4% was about 100-fold more lethal.
2. New risks for infection have emerged. The highest incidence of infection was in the age category 10-19 years, and most cases, by far, occurred in persons < 50 years of age. In addition to young age as a risk factor, other important and somewhat surprising risks for poor outcome were obesity and pregnancy. For instance, the odds ratio for death among patients with morbid obesity (body mass index > 40 kg/m2) was 7.6.[1] Among pregnant women, the odds ratio for admission to the intensive care unit was 7.4 for all pregnant women and 13.2 for those in the third trimester.[2]
3. The mortality data are deceptive. Total mortality estimated by the Centers for Disease Control and Prevention (CDC) was approximately 12,300 persons in the United States. As noted, this flu seemed to displace seasonal influenza almost completely, and the anticipated annual mortality rate associated with seasonal flu is 36,000. Thus, a glance at the numbers appears to show a great advantage for the pandemic (H1N1). However, with seasonal flu, 90% of lethal cases are in persons > 65 years of age; in pandemic flu for 2009, almost 90% of deaths were in persons < 65 years of age. Analysis of these data by life-years lost indicates that this pandemic influenza was substantially worse than most flu seasons.[3]
4. Universal vaccine for influenza is beneficial. A series of reports by Kwong and colleagues from Ontario, Canada compared outcomes of universal influenza vaccination in Ontario with those in Canadian provinces that did not have recommendations for universal vaccination.[4-6] The team from Ontario documented a reduction in illness that translated to 144,000 fewer cases by analyzing the number of respiratory antibiotic prescriptions before and after institution of universal vaccination.[4] Further research on this project indicated a 40%-70% reduction in rates of mortality, hospitalization, use of the emergency department, and physician visits.[5] Economic analyses showed that universal vaccination was also highly cost-effective.[6] This experience is particularly important in view of the recent recommendation by the CDC for influenza vaccine for all persons in the United States > 6 months of age. It must be emphasized that this simply means that vaccination is recommended, but it is up to us to make it work. Influenza vaccination must be readily available in malls, work places, and pharmacies, and it must be cheap or free.
5. Healthcare workers need to receive influenza vaccination. The incredibly embarrassing record of healthcare workers (HCWs) in getting influenza vaccination has been discussed often. In most years, the rate of HCW vaccination averages about 45%-50%, but for the 2009 H1N1 strain the national rate for HCWs was only 36%. The issue of mandatory vaccination for HCWs was also controversial. One healthcare system (BJC HealthCare, St. Louis, Missouri) found that in most seasons, only 32%-54% of HCWs in their system received the influenza vaccine, so BJC made it mandatory as a contingency of employment. With this policy, a vaccination rate for 25,980 HCWs of 98.4% was reported; 0.4% had religious exemptions, 1.2% had medical exemptions, and 8 were fired.[7]
6. The surgical mask wins (maybe). The continuing debate about the relative merits of the N95 respirator mask versus the standard surgical mask to prevent transmission of influenza virus among HCWs was tested in Canada.[8] The study included 478 nurses who were randomly assigned to use either the N95 mask or a conventional surgical mask. The nurses were monitored for evidence of influenza or other viral respiratory tract infections. Of nurses with complete follow-up, the results for surgical masks showed infections in 50 of 212 (24%) compared with 48 of 210 (23%) in nurses who wore fit-tested N95 masks. The investigators concluded that the surgical and N95 masks are equally effective, but surgical masks were also cheap, comfortable, and in great supply. More recently, the Society for Hospital Epidemiology of America (SHEA) at their April 2010 meeting voted to recommend surgical masks.[9] (Note: the CDC still seems to favor the N95 fit-tested masksthat we all hate, but this might change).
7. The epidemic can be tracked with the Internet. The New England Journal of Medicine developed a somewhat novel method of tracking the epidemic of influenza around the world by using news reports and health reports from diverse sources.[10] Some 87,000 sources of information were "filtered" for validity and then used to display a time sequence for global dissemination of influenza, which was readily available to anyone with Internet access. It should be noted that Google did something similar by identifying and tracking the keywords used by consumers seeking information about influenza-like illness. Google was then able to map the US epidemic in real time, and could even predict epidemics about 2 weeks earlier than the CDC.[11]
8. Vaccine production needs to be improved. The current vaccine production system requires eggs and takes 6 months. This year's epidemic was a painful example of our current limitations in vaccine production capacity and speed -- the promised large supply of vaccine did not arrive until the second peak had already started to decline. New technologies are now being pursued that will shorten the production time and magnify the yield by using molecular techniques.[12]
9. Diagnostic testing faces limitations. Polymerase chain reaction has now replaced culture as the gold standard for influenza virus detection. The rapid test used in emergency departments and clinic settings has good specificity but sensitivity of only 60%-70%; thus, a negative test result does not exclude the diagnosis of influenza. This point is emphasized by the many diagnostic errors that resulted in withholding treatment from some patients who needed it.[13]
10. How did the 2009 pandemic (H1N1) kill patients? Early reports showed that primary influenza pneumonia caused by 2009 pandemic (H1N1) histopathologically resembled the highly fatal avian (H5N1) infection.[14] The pathology of 2009 H1N1 (diffuse alveolar damage, intra-alveolar hemorrhage, and the detection of viral antigens within pneumocytes) is quite different from what is found after death from the usual seasonal flu. The other major factor that contributed to pulmonary failure with 2009 (H1N1) was bacterial infection, which was found by the CDC by using a special stain technique, in about one third of fatal cases.[15] Less surprising was the fact that some of the same pathogens found with bacterial infections in the pandemic of 1918-1919 (Streptococcus pneumoniae, Staphylococcus aureus, and group A streptococci) were also found with 2009 (H1N1) influenza-associated infections.[16] These are treatable pathogens, a fact that is important to remember in seriously ill patients.
Summary
The experience with the 2009 Influenza H1N1 pandemic may be the most instructive flu season we have had in decades. Almost unprecedented engagement of the public was evident: public health officials, government, science, and industry. Many will count and lament the failures, but the truth is that we are likely to be much better prepared to deal with multiple influenza-related issues in the future. These include issues about masks, surge capacity, universal vaccine, use of Internet for communications and epidemiology, diagnostic testing, and school closing as an attempt (although failed) to control influenza. What we still don't know is how to get people vaccinated, how to optionally use antiviral drugs, how to make a vaccine in large amounts in less than 6 months, and how to define a pandemic.
References
No comments:
Post a Comment