Sherif B. Mossad, MD

Influenza vaccination of residents of long-term care facilities (and of health care workers at these facilities) is critical for the prevention of influenza in this frail population. Detection, chemoprophylaxis, and treatment have limitations. Infection control measures should be in place during and between outbreaks. Acute care facilities such as emergency departments and hospitals can assist by testing residents of long-term care facilities who present with influenza-like illness during seasonal epidemics of influenza, and by notifying the receiving facility if a patient with influenza would be arriving.

THE EXTENT OF THE PROBLEM

From 5% to 20% of the US population, including residents and health care workers in long-term care facilities, are infected with influenza every year.^1,2 The proportion of those infected who develop clinical illness ranges from 40% to 80%. Each influenza illness is associated with an average of 10 days of respiratory sickness, resulting in approximately 3 days of bed confinement or restricted activity. About 30% to 50% of patients with microbiologically confirmed influenza seek medical care, of whom 16% undergo laboratory tests, 17% undergo radiologic tests, and 75% are recommended an over-the-counter drug or are prescribed a medication. Annual influenza-related hospitalizations range from 200,000 to 400,000, depending on seasonal variations in virulence.^2,3 Thus, influenza causes 1.3 hospitalizations per 1,000 people, and 25% of these are in people age 65 and older. In the United States, about 40,000 to 60,000 people die of influenza every year, and 90% of these are age 65 and older.⁴

POLICIES TO FIGHT ANTIVIRAL RESISTANCE

In January 2006, the US Centers for Disease Control and Prevention (CDC) recommended against the use of the adamantanes—ie, amantadine (Symmetrel) and rimantadine (Flumadine)—for the treatment or prevention of influenza because of a high level of resistance in circulating influenza A/H3N2 in the community. Unfortunately, this resistance trend has not reversed since then, with 96% to 100% of influenza A/H3N2 isolates in the United States showing resistance.⁵ During the 2007–2008 influenza season, influenza A/H1N1 isolates resistant to oseltamivir (Tamiflu) emerged in Europe, particularly in Norway and France. In the United States, influenza A/H1N1 resistance to oseltamivir increased from 0.7% in the 2006–2007 season, to 10.9% in the 2007–2008 season, and to 98% during the 2008–2009 season.^6,7

Fortunately, all oseltamivir-resistant isolates remain susceptible to zanamivir (Relenza). In April 2009, a new influenza A/H1N1 variant (previously referred to as swine-origin influenza virus, or SOIV) emerged in North America and spread to many countries worldwide, and the World Health Organization eventually declared a pandemic. This new variant is susceptible to oseltamivir and zanamivir but resistant to the adamantanes. Resistance patterns for future influenza seasons cannot be predicted, but the current extent of influenza resistance and its development over the past decade⁸ are alarming.

WHY IS INFLUENZA MORE SERIOUS IN LONG-TERM CARE RESIDENTS?

Influenza is usually introduced to long-term care facilities by workers and visitors. Inside, the closed environment and limited mobility of residents facilitate transmission of infection.

The clinical presentation of influenza in residents of long-term care facilities can be subtle, with a blunted febrile response and a decline in mental and functional status.⁹

Residents commonly have underlying diseases that can be exacerbated by influenza infection, such as congestive heart failure, chronic obstructive lung disease, chronic kidney disease, and dementia. In addition, residents are at higher risk of serious influenza-related complications than are community-dwelling elderly people.

Impaired oral intake, limited dexterity, and altered consciousness may limit treatment options, thus further adversely affecting outcomes. Bacterial pneumonia secondary to influenza has dire consequences in long-term care residents. Rates of hospitalization for pneumonia and influenza and for exacerbation of chronic lung disease are higher in these patients than in their community-dwelling counterparts.¹⁰ Death rates are also higher, exceeding 5% during influenza epidemics.¹¹

PREVENTING INFLUENZA IN LONG-TERM CARE FACILITIES

Immunizing residents is essential

Vaccination is the most important measure in preventing influenza in long-term care facilities, and vaccination programs should include residents and health care workers.

Influenza outbreaks are more common in facilities where the rate of immunization is below 80%, as well as in larger facilities (with > 100 beds), suggesting that herd immunity may play a role.¹² Unfortunately, influenza vaccine coverage rates vary widely in this patient population—from 57% to 98% in one report.¹³

The live-attenuated intranasal vaccine is approved only for healthy people under age 50, so most residents of long-term care facilities should receive only the inactivated trivalent intramuscular vaccine.

The effectiveness of a vaccine in preventing influenza depends in part on the adequacy of the match between vaccine serotypes and circulating strains. Studies of all types—randomized, observational, case-control, and cohort studies, as well as meta-analyses and systematic reviews—have shown preventive efficacy rates of influenza vaccination in elderly residents of long-term care facilities to be 23% to 43% for influenza-like illness, 0% to 58% for influenza, 46% for pneumonia, 45% for hospitalization, 42% for death from influenza or pneumonia, and 60% for death from all causes.^14,15 Vaccine performance improved after adjustment for confounders.

Obviously, this protection is variable and incomplete, since influenza outbreaks continue to occur even in long-term care facilities in which most residents are vaccinated.¹³

 

 

Vaccination works, despite the controversy

Whether influenza vaccination prevents deaths in elderly people—or how many deaths it prevents—is a subject of ongoing controversy. ¹⁶ Even though influenza vaccination coverage in the elderly increased from 15% to 65% since 1980, the specific influenza-related death rate did not decrease.¹⁷

It has been suggested that cohort studies may have overestimated the mortality benefit of influenza vaccination in the elderly because of “frailty selection bias” (ie, extremely frail elderly patients are less likely to be vaccinated and are more likely to die for reasons other than influenza than are less frail, vaccinated elderly people) and because of the use of nonspecific end points such as all-cause mortality. ¹⁶ Similarly, observational studies may have overestimated the in-hospital mortality benefit of influenza vaccination in older patients with pneumonia occurring outside of influenza season because of the “healthy user effect” (ie, residual confounding by functional and socioeconomic status).¹⁸

One nested case-control study in immunocompetent elderly patients showed that influenza vaccination was not associated with a reduced risk of community-acquired pneumonia after adjusting for the presence and severity of comorbidities.¹⁹

Since death is a rare end point, it is hard to show a reduction in the death rate with vaccination in randomized controlled studies. The absolute risk reduction in hospitalization and death with vaccination is two to five times higher in elderly patients at high risk than in the healthy elderly.²⁰

The mortality benefit in elderly patients is increased with annual revaccination, with one death prevented for every 300 vaccinations, and one for every 200 revaccinations.²¹

The response to influenza vaccination is reduced in elderly people because of immune senescence, and higher doses of vaccine have been shown to be more immunogenic and remain safe.²² This enhanced antibody response may be maintained for subsequent antigenically different influenza variants, even against viruses appearing more than 10 years after vaccination. ²³ The 2008–2009 influenza vaccine does not protect against the new, pandemic influenza A/H1N1 variant; efforts to produce such a vaccine are under way.

Influenza vaccination is safe. Recent data showed no association between immunization and Guillain-Barré syndrome.²⁴ In fact, influenza itself may be a triggering agent for Guillain-Barré syndrome during major influenza outbreaks.²⁵

DURATION OF SEROPROTECTION IS MORE THAN 6 MONTHS

Every effort should be made to vaccinate residents of long-term care facilities and their caregivers as early as possible in the influenza season to allow an adequate antibody response to develop before the onset of an influenza outbreak.

In the past, there has been concern that the influenza-vaccine-induced antibody response declines more rapidly in the elderly and may fall below seroprotective levels within 4 months of vaccination. But a recent review of 14 published studies argued against that notion, showing that if seroprotection is achieved in the first month after vaccination, it is then maintained for more than 6 months.²⁶ That review also showed that seroconversion varies inversely with preimmunization titers, but not with age.

Moreover, a prospective study²⁷ in 303 residents of a long-term care facility reported that seroprotection did not correlate with nutritional status. In the study, vaccination was very effective despite a high prevalence of nutritional deficiencies. This study also indicated that although an influenza antibody titer greater than 1:40 is considered protective in the general population, long-term care facility residents may require higher levels for effective immunization.

A recent survey showed that a national shortage of influenza vaccine results in decreased immunization rates in residents and in health care workers in long-term care facilities. ²⁸ In that survey, only 2.3% of facilities expressed concern about emergency preparedness, and this has significant implications for a possible influenza pandemic.

VACCINATION PROGRAMS

Standing-order programs have been shown to significantly increase vaccination rates in ambulatory and hospital settings. However, a recent survey showed that only 9% of long-term care facilities use such programs.²⁹ The greatest use of such programs was in government-owned, nonprofit, dual-certified (ie, by both Medicare and Medicaid), and independent long-term care facilities, and in facilities with a lower index of disease acuity. Use varied substantially by state.

The Healthy People 2010 goal of 90% vaccination may be attained by implementing written protocols for documenting immunization—and refusal of immunization—in a consistent place in the patient’s medical record.³⁰

VACCINATION OF WORKERS

When health care workers in long-term care facilities are vaccinated against influenza, significantly fewer residents die^31,32 or develop influenza-like illness, particularly when residents themselves are vaccinated.³³ Additional benefits include decreased need for consultations with general practitioners or admissions to the hospital for influenza-like illness during periods of moderate influenza activity.³²

The policy of mandatory influenza vaccination for health care workers has its proponents³⁴ and opponents.³⁵ When a large tertiary care center adopted mandatory vaccination, vaccination rates increased significantly over time.³⁶ The Centers for Medicare & Medicaid Services recently mandated public reporting of vaccination rates in health care workers and residents of long-term care facilities, and compliance is expected to increase as a result.

BETWEEN INFLUENZA OUTBREAKS

Studies show that hygienic measures prevent transmission of respiratory viruses.³⁷ Therefore, the cornerstones of any program to prevent transmission of influenza and other microorganisms in long-term care facilities are hand hygiene, cough and sneeze etiquette, maintaining a distance of 3 feet between beds, and education of residents and health care workers.

 

 

Wash your hands!

Hands should be washed before and after direct contact with patients or with inanimate objects in their immediate vicinity.³⁸ Contrary to popular belief, hands should also be washed before and after wearing gloves, particularly when handling an invasive device for patient care, and after contact with bodily fluids, excretions, mucous membranes, non-intact skin, or wound dressings. Hands should be washed not only between patients but also during care for the same patient if moving from a contaminated body site to a clean site.

Hands should be washed for 15 to 20 seconds using soap and water or an alcohol-based foam or gel; when hands are visibly soiled, only soap and water should be used. It is not known whether adding virucidals or antiseptics to normal hand-washing further decreases the spread of these viruses.

Continuous education, feedback interventions, and patient-awareness programs can improve hand-washing compliance, but they are not sufficient. Easily accessible dispensers for alcohol-based waterless antiseptic foam or gel can significantly improve hand hygiene rates among health care workers.³⁹

DETECTING INFLUENZA OUTBREAKS

Direct fluorescent antigen detection and nucleic acid detection by polymerase chain reaction (PCR) are the tests recommended for early detection of influenza outbreaks in long-term care facilities. Rapid point-of-care tests to detect influenza antigen are only 60% to 70% sensitive,^13,40 viral culture takes several days, and serologic diagnosis requires documentation of seroconversion at least 2 weeks apart, and so none of these is adequate for the early detection of an influenza outbreak.

There is no widely available test to differentiate influenza A/H1N1 from A/H3N2, or to test for drug resistance. In addition, the PCR tests widely used today do not differentiate between seasonal influenza A/H1N1 and the new, pandemic influenza A/H1N1 variant. Efforts are under way to produce and distribute such a test.

Controlling influenza outbreaks

An outbreak should be declared when two or more residents develop an influenza-like illness within 72 hours of each other during the influenza season.⁴¹ After influenza infection is confirmed by laboratory testing, testing of all residents who subsequently develop an influenza-like illness may not be feasible, and other respiratory viruses may be responsible for mixed outbreaks during influenza epidemics, particularly respiratory syncytial virus.

Roommates of residents who test positive for influenza have a risk of acquiring influenza three times higher than that of residents living in single rooms.⁴² Obviously, private rooms for all residents would be optimal, but this is not practical in most facilities. “Cohorting” (ie, housing infected residents together) is reasonable, but in this situation they might infect each other with other viruses or with influenza viruses of different serotypes.

When an outbreak occurs, potential solutions include closing subunits of the facility to new admissions, limiting movement of residents and health care workers from affected to unaffected units, and using curtains or other barriers between roommates. But most important during seasonal outbreaks are hand hygiene, proper cough and sneeze etiquette, and 3-foot separation between roommates. A simulation model showed that during an influenza pandemic, preventing ill residents of long-term care facilities from making contact with other residents may reduce rates of illness and death by about 60%.⁴³ If a long-term care facility resident is visibly coughing and cannot cover his or her mouth, health care workers should wear a mask when within 3 feet of the patient or when entering the room of a resident with confirmed influenza.

CHEMOPROPHYLAXIS DURING INFLUENZA OUTBREAKS

When influenza outbreaks are recognized early in long-term care facilities, appropriate infection control measures and chemoprophylaxis can be started. Chemoprophylaxis should also be considered in residents in whom influenza vaccination is contraindicated, such as those with severe egg allergy.

The choice of prophylactic agent (Table 1) should be based on the influenza serotype causing the outbreak, if known. If the serotype is influenza A/H1N1, the choice is zanamivir or the combination of oseltamivir plus an adamantane. For influenza A/H3N2 or influenza B, the choice is zanamivir or oseltamivir. Since most clinical laboratories do not offer testing for antiviral resistance or subtyping of influenza A, either zanamivir alone or oseltamivir plus an adamantane is recommended for influenza A.⁷ Experiments in animals support combination therapy, but data in humans are sparse. If available, surveillance data for the predominant epidemic influenza subtype may further guide the choice of prophylactic agent.

Rimantadine is preferred over amantadine in residents of long-term care facilities, since amantadine is associated with a much higher rate of adverse events (18.6% vs 1.9%), especially confusion (10.6% vs 0.6%), resulting in more frequent discontinuation (17.3% vs 1.9%).⁴⁴ In addition, viral resistance to amantadine develops in about 30% of those who receive it. In long-term care facilities, viral resistance occurs not only when it is used for the management of influenza, but also when it is used to treat Parkinson disease.⁴⁵

Oseltamivir prophylaxis for 6 weeks in influenza-vaccinated, frail elderly residents of long-term care facilities during influenza epidemics was 91% effective in preventing laboratory-confirmed clinical influenza, and 85% effective in preventing a secondary bacterial complication such as pneumonia or sinusitis.⁴⁶ The rate of adverse events associated with oseltamivir in that population was similar to that with placebo. Importantly, oseltamivir was not associated with suppression of antibody response to influenza infection or vaccination.

Oseltamivir is very effective in terminating influenza outbreaks in long-term care facilities, even when amantadine fails.^4,13 When used in that manner, it is also associated with decreased antibiotic prescriptions, hospitalizations, deaths,⁴⁷ and substantial cost-savings, even when compared with amantadine, which has a much lower acquisition cost but a higher rate of adverse events, lower efficacy, and individualized dosing requirements.⁴⁸

Zanamivir prophylaxis for 2 weeks given to unvaccinated residents was 29% effective in preventing all symptomatic influenza confirmed by laboratory testing (by culture, PCR, or seroconversion). It was 65% effective in preventing symptomatic culture-confirmed influenza, 70% effective in preventing febrile, laboratory-confirmed influenza, and 21% effective in preventing complications.⁴⁹ It was well tolerated in this population and was not associated with the emergence of zanamivir resistance. Zanamivir also provides 61% additional protective efficacy over rimantadine in vaccinated residents of long-term care facilities,⁵⁰ primarily because of the emergence of viruses resistant to rimantadine. However, up to 50% of elderly people may have difficulty loading and priming the Diskhaler device used to deliver zanamivir.⁵¹

While several studies have shown chemoprophylaxis to be effective, it is not possible to ascribe the decrease in cases during an outbreak entirely to antiviral drugs since most studies were not placebo-controlled, and since the natural tendency of outbreaks is to subside.

 

TREATMENT OF INFLUENZA IN LONG-TERM CARE FACILITIES

The algorithm shown in Figure 1 provides a guide for treatment of an influenza-like illness based on recent recommendations by the CDC⁷ and on subsequent updates during the pandemic. At the time of this writing, these recommendations apply to both seasonal influenza A/H1N1 and the new, pandemic influenza A/H1N1. If a test becomes widely available that can rapidly identify or detect resistance to influenza A/H3N2, seasonal influenza A/H1N1, and pandemic influenza A/H1N1, the recommendations on the use of antiviral agents may be further refined based on more specific data.

A mild illness in someone without a high-risk underlying disease can be managed with symptomatic measures alone, and testing for influenza is at the discretion of the caregiver. Patients who develop a moderate or severe influenza-like illness during an influenza epidemic and who have a high-risk underlying disease or who require hospitalization should be tested for influenza by a sensitive test such as direct fluorescent antigen testing or PCR and should be started on empiric anti-influenza therapy.⁴¹

Most elderly residents of long-term care facilities do have comorbid conditions and are thus at high risk of complications of influenza. During influenza outbreaks, clinicians caring for these patients—and for similar patients in outpatient settings, emergency departments, and acute-care hospitals—should consider testing for influenza, even if the patient has only a mild influenza-like illness.

Of note, no randomized treatment study has been done specifically in residents of long-term care facilities.⁵² Pooled analysis of data from 321 patients at high risk (76 were age 65 or older) in zanamivir treatment studies showed that those who received the drug were sick for 2.5 fewer days than those who received placebo.⁵³ In addition, zanamivir recipients returned to normal activities 3 days sooner, had an 11% reduction in the median total symptom score over 1 to 5 days, and had a rate of complications requiring antibiotics 43% less than placebo recipients. Rates of adverse events were similar in both groups.

A multicenter, randomized, open-label, controlled trial of oseltamivir in the treatment of influenza in high-risk Chinese patients with chronic respiratory diseases (chronic bronchitis, obstructive emphysema, bronchial asthma, or bronchiectasis) or chronic cardiac disease showed that oseltamivir significantly reduced the duration of influenza symptoms by 36.8%, the severity of symptoms by 43.1%, the duration of fever by 45.2%, the time to return to baseline health status by 5 days, and the need for antibiotics, without increasing the total cost of medical care.⁵⁴

Oseltamivir-resistant isolates of seasonal influenza A/H1N1 do not cause different or more severe symptoms than do oseltamivir-susceptible isolates, and they remain susceptible to zanamivir and the adamantanes. There is no widely available test to differentiate influenza A/H1N1 from A/H3N2 or to test for drug resistance. Influenza B and A/H3N2 remain susceptible to both of the neuraminidase inhibitors, ie, oseltamivir and zanamivir.

All currently circulating influenza A/H3N2 isolates remain resistant to the adamantanes. Influenza B is intrinsically resistant to the adamantanes, and some data show that oseltamivir is less effective against influenza B than against influenza A.

Zanamivir is not available in many pharmacies, is not recommended in patients with underlying reactive airway disease, and requires dexterity for administration.

Amantadine is more likely to cause confusion in the elderly than rimantadine, and it is more susceptible to treatment-emergent antiviral resistance.

No parenteral anti-influenza drug is available, but zanamivir and the experimental drug peramivir are undergoing study for parenteral use. An inhaled, long-acting neuraminidase inhibitor, CS-8958, is currently under study.⁵⁵ Other agents currently under development include T-705, a polymerase inhibitor, and DAS181, an attachment inhibitor.

Complementary and alternative therapies for influenza are not established. A recent review of 14 randomized, controlled trials⁵⁶ revealed that the evidence is weak and limited by small sample sizes, poor methodologic quality, or clinically irrelevant effect sizes.

Influenza vaccination of residents of long-term care facilities (and of health care workers at these facilities) is critical for the prevention of influenza in this frail population. Detection, chemoprophylaxis, and treatment have limitations. Infection control measures should be in place during and between outbreaks. Acute care facilities such as emergency departments and hospitals can assist by testing residents of long-term care facilities who present with influenza-like illness during seasonal epidemics of influenza, and by notifying the receiving facility if a patient with influenza would be arriving.

THE EXTENT OF THE PROBLEM

From 5% to 20% of the US population, including residents and health care workers in long-term care facilities, are infected with influenza every year.^1,2 The proportion of those infected who develop clinical illness ranges from 40% to 80%. Each influenza illness is associated with an average of 10 days of respiratory sickness, resulting in approximately 3 days of bed confinement or restricted activity. About 30% to 50% of patients with microbiologically confirmed influenza seek medical care, of whom 16% undergo laboratory tests, 17% undergo radiologic tests, and 75% are recommended an over-the-counter drug or are prescribed a medication. Annual influenza-related hospitalizations range from 200,000 to 400,000, depending on seasonal variations in virulence.^2,3 Thus, influenza causes 1.3 hospitalizations per 1,000 people, and 25% of these are in people age 65 and older. In the United States, about 40,000 to 60,000 people die of influenza every year, and 90% of these are age 65 and older.⁴

POLICIES TO FIGHT ANTIVIRAL RESISTANCE

In January 2006, the US Centers for Disease Control and Prevention (CDC) recommended against the use of the adamantanes—ie, amantadine (Symmetrel) and rimantadine (Flumadine)—for the treatment or prevention of influenza because of a high level of resistance in circulating influenza A/H3N2 in the community. Unfortunately, this resistance trend has not reversed since then, with 96% to 100% of influenza A/H3N2 isolates in the United States showing resistance.⁵ During the 2007–2008 influenza season, influenza A/H1N1 isolates resistant to oseltamivir (Tamiflu) emerged in Europe, particularly in Norway and France. In the United States, influenza A/H1N1 resistance to oseltamivir increased from 0.7% in the 2006–2007 season, to 10.9% in the 2007–2008 season, and to 98% during the 2008–2009 season.^6,7

Fortunately, all oseltamivir-resistant isolates remain susceptible to zanamivir (Relenza). In April 2009, a new influenza A/H1N1 variant (previously referred to as swine-origin influenza virus, or SOIV) emerged in North America and spread to many countries worldwide, and the World Health Organization eventually declared a pandemic. This new variant is susceptible to oseltamivir and zanamivir but resistant to the adamantanes. Resistance patterns for future influenza seasons cannot be predicted, but the current extent of influenza resistance and its development over the past decade⁸ are alarming.

WHY IS INFLUENZA MORE SERIOUS IN LONG-TERM CARE RESIDENTS?

Influenza is usually introduced to long-term care facilities by workers and visitors. Inside, the closed environment and limited mobility of residents facilitate transmission of infection.

The clinical presentation of influenza in residents of long-term care facilities can be subtle, with a blunted febrile response and a decline in mental and functional status.⁹

Residents commonly have underlying diseases that can be exacerbated by influenza infection, such as congestive heart failure, chronic obstructive lung disease, chronic kidney disease, and dementia. In addition, residents are at higher risk of serious influenza-related complications than are community-dwelling elderly people.

Impaired oral intake, limited dexterity, and altered consciousness may limit treatment options, thus further adversely affecting outcomes. Bacterial pneumonia secondary to influenza has dire consequences in long-term care residents. Rates of hospitalization for pneumonia and influenza and for exacerbation of chronic lung disease are higher in these patients than in their community-dwelling counterparts.¹⁰ Death rates are also higher, exceeding 5% during influenza epidemics.¹¹

PREVENTING INFLUENZA IN LONG-TERM CARE FACILITIES

Immunizing residents is essential

Vaccination is the most important measure in preventing influenza in long-term care facilities, and vaccination programs should include residents and health care workers.

Influenza outbreaks are more common in facilities where the rate of immunization is below 80%, as well as in larger facilities (with > 100 beds), suggesting that herd immunity may play a role.¹² Unfortunately, influenza vaccine coverage rates vary widely in this patient population—from 57% to 98% in one report.¹³

The live-attenuated intranasal vaccine is approved only for healthy people under age 50, so most residents of long-term care facilities should receive only the inactivated trivalent intramuscular vaccine.

The effectiveness of a vaccine in preventing influenza depends in part on the adequacy of the match between vaccine serotypes and circulating strains. Studies of all types—randomized, observational, case-control, and cohort studies, as well as meta-analyses and systematic reviews—have shown preventive efficacy rates of influenza vaccination in elderly residents of long-term care facilities to be 23% to 43% for influenza-like illness, 0% to 58% for influenza, 46% for pneumonia, 45% for hospitalization, 42% for death from influenza or pneumonia, and 60% for death from all causes.^14,15 Vaccine performance improved after adjustment for confounders.

Obviously, this protection is variable and incomplete, since influenza outbreaks continue to occur even in long-term care facilities in which most residents are vaccinated.¹³

 

 

Vaccination works, despite the controversy

Whether influenza vaccination prevents deaths in elderly people—or how many deaths it prevents—is a subject of ongoing controversy. ¹⁶ Even though influenza vaccination coverage in the elderly increased from 15% to 65% since 1980, the specific influenza-related death rate did not decrease.¹⁷

It has been suggested that cohort studies may have overestimated the mortality benefit of influenza vaccination in the elderly because of “frailty selection bias” (ie, extremely frail elderly patients are less likely to be vaccinated and are more likely to die for reasons other than influenza than are less frail, vaccinated elderly people) and because of the use of nonspecific end points such as all-cause mortality. ¹⁶ Similarly, observational studies may have overestimated the in-hospital mortality benefit of influenza vaccination in older patients with pneumonia occurring outside of influenza season because of the “healthy user effect” (ie, residual confounding by functional and socioeconomic status).¹⁸

One nested case-control study in immunocompetent elderly patients showed that influenza vaccination was not associated with a reduced risk of community-acquired pneumonia after adjusting for the presence and severity of comorbidities.¹⁹

Since death is a rare end point, it is hard to show a reduction in the death rate with vaccination in randomized controlled studies. The absolute risk reduction in hospitalization and death with vaccination is two to five times higher in elderly patients at high risk than in the healthy elderly.²⁰

The mortality benefit in elderly patients is increased with annual revaccination, with one death prevented for every 300 vaccinations, and one for every 200 revaccinations.²¹

The response to influenza vaccination is reduced in elderly people because of immune senescence, and higher doses of vaccine have been shown to be more immunogenic and remain safe.²² This enhanced antibody response may be maintained for subsequent antigenically different influenza variants, even against viruses appearing more than 10 years after vaccination. ²³ The 2008–2009 influenza vaccine does not protect against the new, pandemic influenza A/H1N1 variant; efforts to produce such a vaccine are under way.

Influenza vaccination is safe. Recent data showed no association between immunization and Guillain-Barré syndrome.²⁴ In fact, influenza itself may be a triggering agent for Guillain-Barré syndrome during major influenza outbreaks.²⁵

DURATION OF SEROPROTECTION IS MORE THAN 6 MONTHS

Every effort should be made to vaccinate residents of long-term care facilities and their caregivers as early as possible in the influenza season to allow an adequate antibody response to develop before the onset of an influenza outbreak.

In the past, there has been concern that the influenza-vaccine-induced antibody response declines more rapidly in the elderly and may fall below seroprotective levels within 4 months of vaccination. But a recent review of 14 published studies argued against that notion, showing that if seroprotection is achieved in the first month after vaccination, it is then maintained for more than 6 months.²⁶ That review also showed that seroconversion varies inversely with preimmunization titers, but not with age.

Moreover, a prospective study²⁷ in 303 residents of a long-term care facility reported that seroprotection did not correlate with nutritional status. In the study, vaccination was very effective despite a high prevalence of nutritional deficiencies. This study also indicated that although an influenza antibody titer greater than 1:40 is considered protective in the general population, long-term care facility residents may require higher levels for effective immunization.

A recent survey showed that a national shortage of influenza vaccine results in decreased immunization rates in residents and in health care workers in long-term care facilities. ²⁸ In that survey, only 2.3% of facilities expressed concern about emergency preparedness, and this has significant implications for a possible influenza pandemic.

VACCINATION PROGRAMS

Standing-order programs have been shown to significantly increase vaccination rates in ambulatory and hospital settings. However, a recent survey showed that only 9% of long-term care facilities use such programs.²⁹ The greatest use of such programs was in government-owned, nonprofit, dual-certified (ie, by both Medicare and Medicaid), and independent long-term care facilities, and in facilities with a lower index of disease acuity. Use varied substantially by state.

The Healthy People 2010 goal of 90% vaccination may be attained by implementing written protocols for documenting immunization—and refusal of immunization—in a consistent place in the patient’s medical record.³⁰

VACCINATION OF WORKERS

When health care workers in long-term care facilities are vaccinated against influenza, significantly fewer residents die^31,32 or develop influenza-like illness, particularly when residents themselves are vaccinated.³³ Additional benefits include decreased need for consultations with general practitioners or admissions to the hospital for influenza-like illness during periods of moderate influenza activity.³²

The policy of mandatory influenza vaccination for health care workers has its proponents³⁴ and opponents.³⁵ When a large tertiary care center adopted mandatory vaccination, vaccination rates increased significantly over time.³⁶ The Centers for Medicare & Medicaid Services recently mandated public reporting of vaccination rates in health care workers and residents of long-term care facilities, and compliance is expected to increase as a result.

BETWEEN INFLUENZA OUTBREAKS

Studies show that hygienic measures prevent transmission of respiratory viruses.³⁷ Therefore, the cornerstones of any program to prevent transmission of influenza and other microorganisms in long-term care facilities are hand hygiene, cough and sneeze etiquette, maintaining a distance of 3 feet between beds, and education of residents and health care workers.

 

 

Wash your hands!

Hands should be washed before and after direct contact with patients or with inanimate objects in their immediate vicinity.³⁸ Contrary to popular belief, hands should also be washed before and after wearing gloves, particularly when handling an invasive device for patient care, and after contact with bodily fluids, excretions, mucous membranes, non-intact skin, or wound dressings. Hands should be washed not only between patients but also during care for the same patient if moving from a contaminated body site to a clean site.

Hands should be washed for 15 to 20 seconds using soap and water or an alcohol-based foam or gel; when hands are visibly soiled, only soap and water should be used. It is not known whether adding virucidals or antiseptics to normal hand-washing further decreases the spread of these viruses.

Continuous education, feedback interventions, and patient-awareness programs can improve hand-washing compliance, but they are not sufficient. Easily accessible dispensers for alcohol-based waterless antiseptic foam or gel can significantly improve hand hygiene rates among health care workers.³⁹

DETECTING INFLUENZA OUTBREAKS

Direct fluorescent antigen detection and nucleic acid detection by polymerase chain reaction (PCR) are the tests recommended for early detection of influenza outbreaks in long-term care facilities. Rapid point-of-care tests to detect influenza antigen are only 60% to 70% sensitive,^13,40 viral culture takes several days, and serologic diagnosis requires documentation of seroconversion at least 2 weeks apart, and so none of these is adequate for the early detection of an influenza outbreak.

There is no widely available test to differentiate influenza A/H1N1 from A/H3N2, or to test for drug resistance. In addition, the PCR tests widely used today do not differentiate between seasonal influenza A/H1N1 and the new, pandemic influenza A/H1N1 variant. Efforts are under way to produce and distribute such a test.

Controlling influenza outbreaks

An outbreak should be declared when two or more residents develop an influenza-like illness within 72 hours of each other during the influenza season.⁴¹ After influenza infection is confirmed by laboratory testing, testing of all residents who subsequently develop an influenza-like illness may not be feasible, and other respiratory viruses may be responsible for mixed outbreaks during influenza epidemics, particularly respiratory syncytial virus.

Roommates of residents who test positive for influenza have a risk of acquiring influenza three times higher than that of residents living in single rooms.⁴² Obviously, private rooms for all residents would be optimal, but this is not practical in most facilities. “Cohorting” (ie, housing infected residents together) is reasonable, but in this situation they might infect each other with other viruses or with influenza viruses of different serotypes.

When an outbreak occurs, potential solutions include closing subunits of the facility to new admissions, limiting movement of residents and health care workers from affected to unaffected units, and using curtains or other barriers between roommates. But most important during seasonal outbreaks are hand hygiene, proper cough and sneeze etiquette, and 3-foot separation between roommates. A simulation model showed that during an influenza pandemic, preventing ill residents of long-term care facilities from making contact with other residents may reduce rates of illness and death by about 60%.⁴³ If a long-term care facility resident is visibly coughing and cannot cover his or her mouth, health care workers should wear a mask when within 3 feet of the patient or when entering the room of a resident with confirmed influenza.

CHEMOPROPHYLAXIS DURING INFLUENZA OUTBREAKS

When influenza outbreaks are recognized early in long-term care facilities, appropriate infection control measures and chemoprophylaxis can be started. Chemoprophylaxis should also be considered in residents in whom influenza vaccination is contraindicated, such as those with severe egg allergy.

The choice of prophylactic agent (Table 1) should be based on the influenza serotype causing the outbreak, if known. If the serotype is influenza A/H1N1, the choice is zanamivir or the combination of oseltamivir plus an adamantane. For influenza A/H3N2 or influenza B, the choice is zanamivir or oseltamivir. Since most clinical laboratories do not offer testing for antiviral resistance or subtyping of influenza A, either zanamivir alone or oseltamivir plus an adamantane is recommended for influenza A.⁷ Experiments in animals support combination therapy, but data in humans are sparse. If available, surveillance data for the predominant epidemic influenza subtype may further guide the choice of prophylactic agent.

Rimantadine is preferred over amantadine in residents of long-term care facilities, since amantadine is associated with a much higher rate of adverse events (18.6% vs 1.9%), especially confusion (10.6% vs 0.6%), resulting in more frequent discontinuation (17.3% vs 1.9%).⁴⁴ In addition, viral resistance to amantadine develops in about 30% of those who receive it. In long-term care facilities, viral resistance occurs not only when it is used for the management of influenza, but also when it is used to treat Parkinson disease.⁴⁵

Oseltamivir prophylaxis for 6 weeks in influenza-vaccinated, frail elderly residents of long-term care facilities during influenza epidemics was 91% effective in preventing laboratory-confirmed clinical influenza, and 85% effective in preventing a secondary bacterial complication such as pneumonia or sinusitis.⁴⁶ The rate of adverse events associated with oseltamivir in that population was similar to that with placebo. Importantly, oseltamivir was not associated with suppression of antibody response to influenza infection or vaccination.

Oseltamivir is very effective in terminating influenza outbreaks in long-term care facilities, even when amantadine fails.^4,13 When used in that manner, it is also associated with decreased antibiotic prescriptions, hospitalizations, deaths,⁴⁷ and substantial cost-savings, even when compared with amantadine, which has a much lower acquisition cost but a higher rate of adverse events, lower efficacy, and individualized dosing requirements.⁴⁸

Zanamivir prophylaxis for 2 weeks given to unvaccinated residents was 29% effective in preventing all symptomatic influenza confirmed by laboratory testing (by culture, PCR, or seroconversion). It was 65% effective in preventing symptomatic culture-confirmed influenza, 70% effective in preventing febrile, laboratory-confirmed influenza, and 21% effective in preventing complications.⁴⁹ It was well tolerated in this population and was not associated with the emergence of zanamivir resistance. Zanamivir also provides 61% additional protective efficacy over rimantadine in vaccinated residents of long-term care facilities,⁵⁰ primarily because of the emergence of viruses resistant to rimantadine. However, up to 50% of elderly people may have difficulty loading and priming the Diskhaler device used to deliver zanamivir.⁵¹

While several studies have shown chemoprophylaxis to be effective, it is not possible to ascribe the decrease in cases during an outbreak entirely to antiviral drugs since most studies were not placebo-controlled, and since the natural tendency of outbreaks is to subside.

 

TREATMENT OF INFLUENZA IN LONG-TERM CARE FACILITIES

The algorithm shown in Figure 1 provides a guide for treatment of an influenza-like illness based on recent recommendations by the CDC⁷ and on subsequent updates during the pandemic. At the time of this writing, these recommendations apply to both seasonal influenza A/H1N1 and the new, pandemic influenza A/H1N1. If a test becomes widely available that can rapidly identify or detect resistance to influenza A/H3N2, seasonal influenza A/H1N1, and pandemic influenza A/H1N1, the recommendations on the use of antiviral agents may be further refined based on more specific data.

A mild illness in someone without a high-risk underlying disease can be managed with symptomatic measures alone, and testing for influenza is at the discretion of the caregiver. Patients who develop a moderate or severe influenza-like illness during an influenza epidemic and who have a high-risk underlying disease or who require hospitalization should be tested for influenza by a sensitive test such as direct fluorescent antigen testing or PCR and should be started on empiric anti-influenza therapy.⁴¹

Most elderly residents of long-term care facilities do have comorbid conditions and are thus at high risk of complications of influenza. During influenza outbreaks, clinicians caring for these patients—and for similar patients in outpatient settings, emergency departments, and acute-care hospitals—should consider testing for influenza, even if the patient has only a mild influenza-like illness.

Of note, no randomized treatment study has been done specifically in residents of long-term care facilities.⁵² Pooled analysis of data from 321 patients at high risk (76 were age 65 or older) in zanamivir treatment studies showed that those who received the drug were sick for 2.5 fewer days than those who received placebo.⁵³ In addition, zanamivir recipients returned to normal activities 3 days sooner, had an 11% reduction in the median total symptom score over 1 to 5 days, and had a rate of complications requiring antibiotics 43% less than placebo recipients. Rates of adverse events were similar in both groups.

A multicenter, randomized, open-label, controlled trial of oseltamivir in the treatment of influenza in high-risk Chinese patients with chronic respiratory diseases (chronic bronchitis, obstructive emphysema, bronchial asthma, or bronchiectasis) or chronic cardiac disease showed that oseltamivir significantly reduced the duration of influenza symptoms by 36.8%, the severity of symptoms by 43.1%, the duration of fever by 45.2%, the time to return to baseline health status by 5 days, and the need for antibiotics, without increasing the total cost of medical care.⁵⁴

Oseltamivir-resistant isolates of seasonal influenza A/H1N1 do not cause different or more severe symptoms than do oseltamivir-susceptible isolates, and they remain susceptible to zanamivir and the adamantanes. There is no widely available test to differentiate influenza A/H1N1 from A/H3N2 or to test for drug resistance. Influenza B and A/H3N2 remain susceptible to both of the neuraminidase inhibitors, ie, oseltamivir and zanamivir.

All currently circulating influenza A/H3N2 isolates remain resistant to the adamantanes. Influenza B is intrinsically resistant to the adamantanes, and some data show that oseltamivir is less effective against influenza B than against influenza A.

Zanamivir is not available in many pharmacies, is not recommended in patients with underlying reactive airway disease, and requires dexterity for administration.

Amantadine is more likely to cause confusion in the elderly than rimantadine, and it is more susceptible to treatment-emergent antiviral resistance.

No parenteral anti-influenza drug is available, but zanamivir and the experimental drug peramivir are undergoing study for parenteral use. An inhaled, long-acting neuraminidase inhibitor, CS-8958, is currently under study.⁵⁵ Other agents currently under development include T-705, a polymerase inhibitor, and DAS181, an attachment inhibitor.

Complementary and alternative therapies for influenza are not established. A recent review of 14 randomized, controlled trials⁵⁶ revealed that the evidence is weak and limited by small sample sizes, poor methodologic quality, or clinically irrelevant effect sizes.

Influenza vaccination of residents of long-term care facilities (and of health care workers at these facilities) is critical for the prevention of influenza in this frail population. Detection, chemoprophylaxis, and treatment have limitations. Infection control measures should be in place during and between outbreaks. Acute care facilities such as emergency departments and hospitals can assist by testing residents of long-term care facilities who present with influenza-like illness during seasonal epidemics of influenza, and by notifying the receiving facility if a patient with influenza would be arriving.

THE EXTENT OF THE PROBLEM

From 5% to 20% of the US population, including residents and health care workers in long-term care facilities, are infected with influenza every year.^1,2 The proportion of those infected who develop clinical illness ranges from 40% to 80%. Each influenza illness is associated with an average of 10 days of respiratory sickness, resulting in approximately 3 days of bed confinement or restricted activity. About 30% to 50% of patients with microbiologically confirmed influenza seek medical care, of whom 16% undergo laboratory tests, 17% undergo radiologic tests, and 75% are recommended an over-the-counter drug or are prescribed a medication. Annual influenza-related hospitalizations range from 200,000 to 400,000, depending on seasonal variations in virulence.^2,3 Thus, influenza causes 1.3 hospitalizations per 1,000 people, and 25% of these are in people age 65 and older. In the United States, about 40,000 to 60,000 people die of influenza every year, and 90% of these are age 65 and older.⁴

POLICIES TO FIGHT ANTIVIRAL RESISTANCE

In January 2006, the US Centers for Disease Control and Prevention (CDC) recommended against the use of the adamantanes—ie, amantadine (Symmetrel) and rimantadine (Flumadine)—for the treatment or prevention of influenza because of a high level of resistance in circulating influenza A/H3N2 in the community. Unfortunately, this resistance trend has not reversed since then, with 96% to 100% of influenza A/H3N2 isolates in the United States showing resistance.⁵ During the 2007–2008 influenza season, influenza A/H1N1 isolates resistant to oseltamivir (Tamiflu) emerged in Europe, particularly in Norway and France. In the United States, influenza A/H1N1 resistance to oseltamivir increased from 0.7% in the 2006–2007 season, to 10.9% in the 2007–2008 season, and to 98% during the 2008–2009 season.^6,7

Fortunately, all oseltamivir-resistant isolates remain susceptible to zanamivir (Relenza). In April 2009, a new influenza A/H1N1 variant (previously referred to as swine-origin influenza virus, or SOIV) emerged in North America and spread to many countries worldwide, and the World Health Organization eventually declared a pandemic. This new variant is susceptible to oseltamivir and zanamivir but resistant to the adamantanes. Resistance patterns for future influenza seasons cannot be predicted, but the current extent of influenza resistance and its development over the past decade⁸ are alarming.

WHY IS INFLUENZA MORE SERIOUS IN LONG-TERM CARE RESIDENTS?

Influenza is usually introduced to long-term care facilities by workers and visitors. Inside, the closed environment and limited mobility of residents facilitate transmission of infection.

The clinical presentation of influenza in residents of long-term care facilities can be subtle, with a blunted febrile response and a decline in mental and functional status.⁹

Residents commonly have underlying diseases that can be exacerbated by influenza infection, such as congestive heart failure, chronic obstructive lung disease, chronic kidney disease, and dementia. In addition, residents are at higher risk of serious influenza-related complications than are community-dwelling elderly people.

Impaired oral intake, limited dexterity, and altered consciousness may limit treatment options, thus further adversely affecting outcomes. Bacterial pneumonia secondary to influenza has dire consequences in long-term care residents. Rates of hospitalization for pneumonia and influenza and for exacerbation of chronic lung disease are higher in these patients than in their community-dwelling counterparts.¹⁰ Death rates are also higher, exceeding 5% during influenza epidemics.¹¹

PREVENTING INFLUENZA IN LONG-TERM CARE FACILITIES

Immunizing residents is essential

Vaccination is the most important measure in preventing influenza in long-term care facilities, and vaccination programs should include residents and health care workers.

Influenza outbreaks are more common in facilities where the rate of immunization is below 80%, as well as in larger facilities (with > 100 beds), suggesting that herd immunity may play a role.¹² Unfortunately, influenza vaccine coverage rates vary widely in this patient population—from 57% to 98% in one report.¹³

The live-attenuated intranasal vaccine is approved only for healthy people under age 50, so most residents of long-term care facilities should receive only the inactivated trivalent intramuscular vaccine.

The effectiveness of a vaccine in preventing influenza depends in part on the adequacy of the match between vaccine serotypes and circulating strains. Studies of all types—randomized, observational, case-control, and cohort studies, as well as meta-analyses and systematic reviews—have shown preventive efficacy rates of influenza vaccination in elderly residents of long-term care facilities to be 23% to 43% for influenza-like illness, 0% to 58% for influenza, 46% for pneumonia, 45% for hospitalization, 42% for death from influenza or pneumonia, and 60% for death from all causes.^14,15 Vaccine performance improved after adjustment for confounders.

Obviously, this protection is variable and incomplete, since influenza outbreaks continue to occur even in long-term care facilities in which most residents are vaccinated.¹³

 

 

Vaccination works, despite the controversy

Whether influenza vaccination prevents deaths in elderly people—or how many deaths it prevents—is a subject of ongoing controversy. ¹⁶ Even though influenza vaccination coverage in the elderly increased from 15% to 65% since 1980, the specific influenza-related death rate did not decrease.¹⁷

It has been suggested that cohort studies may have overestimated the mortality benefit of influenza vaccination in the elderly because of “frailty selection bias” (ie, extremely frail elderly patients are less likely to be vaccinated and are more likely to die for reasons other than influenza than are less frail, vaccinated elderly people) and because of the use of nonspecific end points such as all-cause mortality. ¹⁶ Similarly, observational studies may have overestimated the in-hospital mortality benefit of influenza vaccination in older patients with pneumonia occurring outside of influenza season because of the “healthy user effect” (ie, residual confounding by functional and socioeconomic status).¹⁸

One nested case-control study in immunocompetent elderly patients showed that influenza vaccination was not associated with a reduced risk of community-acquired pneumonia after adjusting for the presence and severity of comorbidities.¹⁹

Since death is a rare end point, it is hard to show a reduction in the death rate with vaccination in randomized controlled studies. The absolute risk reduction in hospitalization and death with vaccination is two to five times higher in elderly patients at high risk than in the healthy elderly.²⁰

The mortality benefit in elderly patients is increased with annual revaccination, with one death prevented for every 300 vaccinations, and one for every 200 revaccinations.²¹

The response to influenza vaccination is reduced in elderly people because of immune senescence, and higher doses of vaccine have been shown to be more immunogenic and remain safe.²² This enhanced antibody response may be maintained for subsequent antigenically different influenza variants, even against viruses appearing more than 10 years after vaccination. ²³ The 2008–2009 influenza vaccine does not protect against the new, pandemic influenza A/H1N1 variant; efforts to produce such a vaccine are under way.

Influenza vaccination is safe. Recent data showed no association between immunization and Guillain-Barré syndrome.²⁴ In fact, influenza itself may be a triggering agent for Guillain-Barré syndrome during major influenza outbreaks.²⁵

DURATION OF SEROPROTECTION IS MORE THAN 6 MONTHS

Every effort should be made to vaccinate residents of long-term care facilities and their caregivers as early as possible in the influenza season to allow an adequate antibody response to develop before the onset of an influenza outbreak.

In the past, there has been concern that the influenza-vaccine-induced antibody response declines more rapidly in the elderly and may fall below seroprotective levels within 4 months of vaccination. But a recent review of 14 published studies argued against that notion, showing that if seroprotection is achieved in the first month after vaccination, it is then maintained for more than 6 months.²⁶ That review also showed that seroconversion varies inversely with preimmunization titers, but not with age.

Moreover, a prospective study²⁷ in 303 residents of a long-term care facility reported that seroprotection did not correlate with nutritional status. In the study, vaccination was very effective despite a high prevalence of nutritional deficiencies. This study also indicated that although an influenza antibody titer greater than 1:40 is considered protective in the general population, long-term care facility residents may require higher levels for effective immunization.

A recent survey showed that a national shortage of influenza vaccine results in decreased immunization rates in residents and in health care workers in long-term care facilities. ²⁸ In that survey, only 2.3% of facilities expressed concern about emergency preparedness, and this has significant implications for a possible influenza pandemic.

VACCINATION PROGRAMS

Standing-order programs have been shown to significantly increase vaccination rates in ambulatory and hospital settings. However, a recent survey showed that only 9% of long-term care facilities use such programs.²⁹ The greatest use of such programs was in government-owned, nonprofit, dual-certified (ie, by both Medicare and Medicaid), and independent long-term care facilities, and in facilities with a lower index of disease acuity. Use varied substantially by state.

The Healthy People 2010 goal of 90% vaccination may be attained by implementing written protocols for documenting immunization—and refusal of immunization—in a consistent place in the patient’s medical record.³⁰

VACCINATION OF WORKERS

When health care workers in long-term care facilities are vaccinated against influenza, significantly fewer residents die^31,32 or develop influenza-like illness, particularly when residents themselves are vaccinated.³³ Additional benefits include decreased need for consultations with general practitioners or admissions to the hospital for influenza-like illness during periods of moderate influenza activity.³²

The policy of mandatory influenza vaccination for health care workers has its proponents³⁴ and opponents.³⁵ When a large tertiary care center adopted mandatory vaccination, vaccination rates increased significantly over time.³⁶ The Centers for Medicare & Medicaid Services recently mandated public reporting of vaccination rates in health care workers and residents of long-term care facilities, and compliance is expected to increase as a result.

BETWEEN INFLUENZA OUTBREAKS

Studies show that hygienic measures prevent transmission of respiratory viruses.³⁷ Therefore, the cornerstones of any program to prevent transmission of influenza and other microorganisms in long-term care facilities are hand hygiene, cough and sneeze etiquette, maintaining a distance of 3 feet between beds, and education of residents and health care workers.

 

 

Wash your hands!

Hands should be washed before and after direct contact with patients or with inanimate objects in their immediate vicinity.³⁸ Contrary to popular belief, hands should also be washed before and after wearing gloves, particularly when handling an invasive device for patient care, and after contact with bodily fluids, excretions, mucous membranes, non-intact skin, or wound dressings. Hands should be washed not only between patients but also during care for the same patient if moving from a contaminated body site to a clean site.

Hands should be washed for 15 to 20 seconds using soap and water or an alcohol-based foam or gel; when hands are visibly soiled, only soap and water should be used. It is not known whether adding virucidals or antiseptics to normal hand-washing further decreases the spread of these viruses.

Continuous education, feedback interventions, and patient-awareness programs can improve hand-washing compliance, but they are not sufficient. Easily accessible dispensers for alcohol-based waterless antiseptic foam or gel can significantly improve hand hygiene rates among health care workers.³⁹

DETECTING INFLUENZA OUTBREAKS

Direct fluorescent antigen detection and nucleic acid detection by polymerase chain reaction (PCR) are the tests recommended for early detection of influenza outbreaks in long-term care facilities. Rapid point-of-care tests to detect influenza antigen are only 60% to 70% sensitive,^13,40 viral culture takes several days, and serologic diagnosis requires documentation of seroconversion at least 2 weeks apart, and so none of these is adequate for the early detection of an influenza outbreak.

There is no widely available test to differentiate influenza A/H1N1 from A/H3N2, or to test for drug resistance. In addition, the PCR tests widely used today do not differentiate between seasonal influenza A/H1N1 and the new, pandemic influenza A/H1N1 variant. Efforts are under way to produce and distribute such a test.

Controlling influenza outbreaks

An outbreak should be declared when two or more residents develop an influenza-like illness within 72 hours of each other during the influenza season.⁴¹ After influenza infection is confirmed by laboratory testing, testing of all residents who subsequently develop an influenza-like illness may not be feasible, and other respiratory viruses may be responsible for mixed outbreaks during influenza epidemics, particularly respiratory syncytial virus.

Roommates of residents who test positive for influenza have a risk of acquiring influenza three times higher than that of residents living in single rooms.⁴² Obviously, private rooms for all residents would be optimal, but this is not practical in most facilities. “Cohorting” (ie, housing infected residents together) is reasonable, but in this situation they might infect each other with other viruses or with influenza viruses of different serotypes.

When an outbreak occurs, potential solutions include closing subunits of the facility to new admissions, limiting movement of residents and health care workers from affected to unaffected units, and using curtains or other barriers between roommates. But most important during seasonal outbreaks are hand hygiene, proper cough and sneeze etiquette, and 3-foot separation between roommates. A simulation model showed that during an influenza pandemic, preventing ill residents of long-term care facilities from making contact with other residents may reduce rates of illness and death by about 60%.⁴³ If a long-term care facility resident is visibly coughing and cannot cover his or her mouth, health care workers should wear a mask when within 3 feet of the patient or when entering the room of a resident with confirmed influenza.

CHEMOPROPHYLAXIS DURING INFLUENZA OUTBREAKS

When influenza outbreaks are recognized early in long-term care facilities, appropriate infection control measures and chemoprophylaxis can be started. Chemoprophylaxis should also be considered in residents in whom influenza vaccination is contraindicated, such as those with severe egg allergy.

The choice of prophylactic agent (Table 1) should be based on the influenza serotype causing the outbreak, if known. If the serotype is influenza A/H1N1, the choice is zanamivir or the combination of oseltamivir plus an adamantane. For influenza A/H3N2 or influenza B, the choice is zanamivir or oseltamivir. Since most clinical laboratories do not offer testing for antiviral resistance or subtyping of influenza A, either zanamivir alone or oseltamivir plus an adamantane is recommended for influenza A.⁷ Experiments in animals support combination therapy, but data in humans are sparse. If available, surveillance data for the predominant epidemic influenza subtype may further guide the choice of prophylactic agent.

Rimantadine is preferred over amantadine in residents of long-term care facilities, since amantadine is associated with a much higher rate of adverse events (18.6% vs 1.9%), especially confusion (10.6% vs 0.6%), resulting in more frequent discontinuation (17.3% vs 1.9%).⁴⁴ In addition, viral resistance to amantadine develops in about 30% of those who receive it. In long-term care facilities, viral resistance occurs not only when it is used for the management of influenza, but also when it is used to treat Parkinson disease.⁴⁵

Oseltamivir prophylaxis for 6 weeks in influenza-vaccinated, frail elderly residents of long-term care facilities during influenza epidemics was 91% effective in preventing laboratory-confirmed clinical influenza, and 85% effective in preventing a secondary bacterial complication such as pneumonia or sinusitis.⁴⁶ The rate of adverse events associated with oseltamivir in that population was similar to that with placebo. Importantly, oseltamivir was not associated with suppression of antibody response to influenza infection or vaccination.

Oseltamivir is very effective in terminating influenza outbreaks in long-term care facilities, even when amantadine fails.^4,13 When used in that manner, it is also associated with decreased antibiotic prescriptions, hospitalizations, deaths,⁴⁷ and substantial cost-savings, even when compared with amantadine, which has a much lower acquisition cost but a higher rate of adverse events, lower efficacy, and individualized dosing requirements.⁴⁸

Zanamivir prophylaxis for 2 weeks given to unvaccinated residents was 29% effective in preventing all symptomatic influenza confirmed by laboratory testing (by culture, PCR, or seroconversion). It was 65% effective in preventing symptomatic culture-confirmed influenza, 70% effective in preventing febrile, laboratory-confirmed influenza, and 21% effective in preventing complications.⁴⁹ It was well tolerated in this population and was not associated with the emergence of zanamivir resistance. Zanamivir also provides 61% additional protective efficacy over rimantadine in vaccinated residents of long-term care facilities,⁵⁰ primarily because of the emergence of viruses resistant to rimantadine. However, up to 50% of elderly people may have difficulty loading and priming the Diskhaler device used to deliver zanamivir.⁵¹

While several studies have shown chemoprophylaxis to be effective, it is not possible to ascribe the decrease in cases during an outbreak entirely to antiviral drugs since most studies were not placebo-controlled, and since the natural tendency of outbreaks is to subside.

 

TREATMENT OF INFLUENZA IN LONG-TERM CARE FACILITIES

The algorithm shown in Figure 1 provides a guide for treatment of an influenza-like illness based on recent recommendations by the CDC⁷ and on subsequent updates during the pandemic. At the time of this writing, these recommendations apply to both seasonal influenza A/H1N1 and the new, pandemic influenza A/H1N1. If a test becomes widely available that can rapidly identify or detect resistance to influenza A/H3N2, seasonal influenza A/H1N1, and pandemic influenza A/H1N1, the recommendations on the use of antiviral agents may be further refined based on more specific data.

A mild illness in someone without a high-risk underlying disease can be managed with symptomatic measures alone, and testing for influenza is at the discretion of the caregiver. Patients who develop a moderate or severe influenza-like illness during an influenza epidemic and who have a high-risk underlying disease or who require hospitalization should be tested for influenza by a sensitive test such as direct fluorescent antigen testing or PCR and should be started on empiric anti-influenza therapy.⁴¹

Most elderly residents of long-term care facilities do have comorbid conditions and are thus at high risk of complications of influenza. During influenza outbreaks, clinicians caring for these patients—and for similar patients in outpatient settings, emergency departments, and acute-care hospitals—should consider testing for influenza, even if the patient has only a mild influenza-like illness.

Of note, no randomized treatment study has been done specifically in residents of long-term care facilities.⁵² Pooled analysis of data from 321 patients at high risk (76 were age 65 or older) in zanamivir treatment studies showed that those who received the drug were sick for 2.5 fewer days than those who received placebo.⁵³ In addition, zanamivir recipients returned to normal activities 3 days sooner, had an 11% reduction in the median total symptom score over 1 to 5 days, and had a rate of complications requiring antibiotics 43% less than placebo recipients. Rates of adverse events were similar in both groups.

A multicenter, randomized, open-label, controlled trial of oseltamivir in the treatment of influenza in high-risk Chinese patients with chronic respiratory diseases (chronic bronchitis, obstructive emphysema, bronchial asthma, or bronchiectasis) or chronic cardiac disease showed that oseltamivir significantly reduced the duration of influenza symptoms by 36.8%, the severity of symptoms by 43.1%, the duration of fever by 45.2%, the time to return to baseline health status by 5 days, and the need for antibiotics, without increasing the total cost of medical care.⁵⁴

Oseltamivir-resistant isolates of seasonal influenza A/H1N1 do not cause different or more severe symptoms than do oseltamivir-susceptible isolates, and they remain susceptible to zanamivir and the adamantanes. There is no widely available test to differentiate influenza A/H1N1 from A/H3N2 or to test for drug resistance. Influenza B and A/H3N2 remain susceptible to both of the neuraminidase inhibitors, ie, oseltamivir and zanamivir.

All currently circulating influenza A/H3N2 isolates remain resistant to the adamantanes. Influenza B is intrinsically resistant to the adamantanes, and some data show that oseltamivir is less effective against influenza B than against influenza A.

Zanamivir is not available in many pharmacies, is not recommended in patients with underlying reactive airway disease, and requires dexterity for administration.

Amantadine is more likely to cause confusion in the elderly than rimantadine, and it is more susceptible to treatment-emergent antiviral resistance.

No parenteral anti-influenza drug is available, but zanamivir and the experimental drug peramivir are undergoing study for parenteral use. An inhaled, long-acting neuraminidase inhibitor, CS-8958, is currently under study.⁵⁵ Other agents currently under development include T-705, a polymerase inhibitor, and DAS181, an attachment inhibitor.

Complementary and alternative therapies for influenza are not established. A recent review of 14 randomized, controlled trials⁵⁶ revealed that the evidence is weak and limited by small sample sizes, poor methodologic quality, or clinically irrelevant effect sizes.

Last year, some people may have lost their faith in flu shots. The three antigens chosen for the vaccine in advance by the US Centers for Disease Control and Prevention (CDC) did not match very well the influenza strains that ultimately circulated in North America, and the overall protective efficacy of the vaccine was estimated at only 40%.

Nevertheless, vaccination remains the primary preventive measure for both epidemic and pandemic influenza, especially in view of a rising rate of resistance to the oral antiviral agent oseltamivir (Tamiflu).

In the 2008–2009 influenza season, we hope to do better. All three antigens contained in the 2008–2009 vaccine are new. Surveillance data from the Southern Hemisphere during the summer of 2008 show that this vaccine is expected to be a good match for the strains circulating in the Northern Hemisphere. And with 146 million doses expected to be manufactured this season by six companies—the largest number of doses ever manufactured in the United States—enough should be available for all.

GREAT STRIDES HAVE BEEN MADE, BUT FLU IS STILL A PROBLEM

We are making great strides against influenza. Over the last 50 years, the rate of influenzarelated deaths in the United States declined by 95%, from an average seasonal rate of 10.2 deaths per 100,000 population in the 1940s to 0.56 per 100,000 by the 1990s.¹

However, influenza still accounts for about 10% of patients admitted to intensive care units for acute respiratory failure during epidemics.²

Children and the elderly are still hit the hardest: infants age 0 through 23 months and adults age 65 years and older have the highest peak rates of pneumonia and influenza hospitalization and death.³ School-age children (5–18 years) have an indirect role in anticipating the risk to others and can learn to help avoid spreading the virus by washing their hands more, wearing masks, and adopting other hygienic measures.

In the 1918–1919 pandemic, most deaths were from secondary bacterial pneumonia, a fact that has implications for pandemic preparedness. ⁴ Currently, Staphylococcus aureus, particularly methicillin-resistant strains (MRSA), is an important cause of secondary bacterial pneumonia, with a high mortality rate.⁵

UPDATE ON DIAGNOSIS: PCR IS THE BEST TEST

In the hospital, it is important to identify patients who have influenza so that we can give them appropriate antiviral therapy and also protect other patients from getting the flu. Unfortunately, the sensitivity and positive predictive value of fever, cough, and other symptoms for the diagnosis of influenza in hospitalized patients are 40% or less.⁶

Real-time reverse transcriptase polymerase chain reaction (PCR), compared with direct fluorescent antigen detection or cell culture, has the highest sensitivity (98.7%) and specificity (100%) in both children⁷ and the elderly.⁸ Furthermore, cell culture is slow and therefore is not useful in clinical practice. Nasopharyngeal wash sampling appears impractical in nursing home residents, owing to their underlying disabilities, and nasopharyngeal swabs tested by PCR are equally sensitive. ⁸

However, improvements are needed in molecular detection and subtyping of influenza viruses.⁹ If a pandemic breaks out, we will need to identify the virus quickly to have enough time for preventive interventions. The US Food and Drug Administration has recently cleared a new test called the Human Virus Real-Time RT-PCR Detection and Characterization Panel to detect and differentiate between seasonal and novel influenza strains.¹⁰

UPDATE ON INFLUENZA VACCINE

New recommendations in 2008 by the CDC Advisory Committee on Immunization Practices¹¹ include annual vaccination for all children age 5 through 18 years, and either the trivalent inactivated vaccine (ie, the shot) or the live-attenuated vaccine (ie, the Flu-Mist intranasal spray) for healthy people age 2 through 49 years. The CDC recommendations are summarized at www.cdc.gov/flu/professionals/acip/index.htm.

Has the benefit of vaccination in adults been overestimated?

Jackson et al,¹² in an article published in August 2008, suggested that the effect of influenza vaccination on the risk of community-acquired pneumonia in immunocompetent elderly people during influenza season is less than previously estimated. However, some patients in this study who were classified as not having been vaccinated may have actually been vaccinated by other health care providers without notifying their primary care providers. Moreover, influenza infection may cause only a small proportion of cases of pneumonia in this population.

In another study, Eurich et al¹³ suggested that previous observational studies overestimated the benefit of influenza vaccination on reducing deaths in patients with pneumonia outside the flu season. Although they found the incidence of death to be 51% lower in vaccinated than in unvaccinated adults with community-acquired pneumonia (N = 1,813) admitted to six hospitals, they ascribed it to confounding factors, specifically socioeconomic and functional status. This phenomenon was previously called the “frailty bias” or the “healthy user effect.” However, this study included only patients hospitalized with pneumonia and did not include data on vaccine-induced immunity or the cause of pneumonia, and measures of the healthy user effect were rudimentary. In addition, only outcomes during hospitalization were included.

Most experts still believe that vaccination prevents 50% of influenza-related deaths (with a smaller effect on rates of all-cause mortality¹⁴), including deaths in very old people. ¹⁵ A recent review found no basis for the historic concern that the antibody response to the influenza vaccine in people age 60 and older declines more rapidly than in younger people and below seroprotective levels within 4 months of immunization.¹⁶

Nevertheless, discordance between antibody and T-cell responses to influenza vaccine does exist¹⁷ (ie, the vaccine can induce antibodies while not boosting the T-cell-specific response), and we should continue to seek new vaccines that are more effective.

 

 

More people are being vaccinated, but we’re still below our goals

Although influenza vaccination rates among adults continue to improve,¹⁸ they remain well below the Healthy People 2010 initiative’s target of 90% in adults age 65 and older (the current rate is 72%) and below the target of 60% in people age 18 through 64 who have one or more high-risk conditions, health care workers, and pregnant women¹⁹ (currently 35% in people age 18 through 49 and 42% in people age 50 through 64). Thus, we still need to improve vaccination coverage rates.

Health care providers should offer vaccination at every opportunity between October and May.²⁰ Offering vaccination in nontraditional settings such as work sites and pharmacies is likely to be cost-saving for healthy adults due to averted morbidity.²¹ At many hospitals, health care workers can opt out of being vaccinated, but they must formally state that they are doing so. The use of these declination statements among health care workers is associated with a mean increase of 11.6% in vaccination rates.²²

Since influenza is the second most frequent vaccine-preventable infection in travelers, the vaccine should be offered to those crossing to the opposite hemisphere during its peak influenza season (eg, to South America in May through September), as well as to those visiting the tropics at any time of year.²³

Vaccination is safe and effective in high-risk groups

Data on vaccination are reassuring in several at-risk groups.

In pregnancy, there is no indication that infants are harmed if their mothers are vaccinated in the first trimester.²⁴ The evidence of excess morbidity during influenza epidemics supports vaccinating healthy pregnant women in the second or third trimester and those with comorbidities any time during pregnancy. Influenza vaccination during pregnancy reduces laboratory-confirmed influenza in infants up to 6 months of age by 63% and prevents 29% of all febrile respiratory illnesses in infants and 36% of those in mothers.²⁵

In patients with chronic obstructive pulmonary disease, vaccination cuts the rate of outpatient visits and hospitalizations due to acute respiratory illness by 67%.²⁶ The antibody response to influenza vaccine in patients with rheumatoid arthritis treated with rituximab (Rituxan), a monoclonal antibody directed against CD20 surface antigen-positive B lymphocytes, was lower than in healthy controls, but was not negligible.²⁷

Dispelling myths about vaccination in children

One recently published study in children younger than 5 years did not find vaccination to be effective in preventing influenza-related hospitalizations and outpatient visits.²⁸ However, in both seasons in which this study was conducted, there was a suboptimal antigenic match between vaccines and circulating strains. Moreover, about 60% of participants were unvaccinated and another 20% were only partially vaccinated, making it difficult to assess vaccine effectiveness. Several other studies have shown that, when there is a good match, vaccine effectiveness in children is 85% to 90%.

Even though the live-attenuated (inhaled) vaccine is more expensive than the inactivated (injected) vaccine, it reduces the number of influenza illness cases and lowers subsequent health care use in children and productivity loss in their parents, with a net total savings of $45.80 relative to the inactivated vaccine.²⁹ The live-attenuated vaccine provides sustained protection against influenza illness for 12 months following vaccination, as well as meaningful efficacy through a second season without revaccination, although at a lower level.³⁰

Several myths about the live-attenuated vaccine should be dispelled.³¹ It is well tolerated and causes only mild, transient symptoms of upper respiratory infection, even in people with asthma or the early stages of human immunodeficiency virus infection. Genetic reversion of the vaccine strain to a wild-type virus requires independent mutation in four gene segments, an event that has not been observed. Finally, although viral shedding is common for several days after vaccination, transmission to another person has been shown in only one person, who remained asymptomatic.

Unfortunately, rates of influenza vaccination are even worse for children than for adults.³² In children 6 through 23 months old, only 22% are fully vaccinated; in those 24 through 59 months old, only 16.5% are.

One group of immunocompromised children, liver transplant recipients, achieved antibody seroprotection and seroconversion rates similar to those achieved by their healthy siblings, with no vaccine-related serious side effects.³³ As in adults, the cell-mediated immune response to the vaccine was diminished, suggesting that other strategies are needed to provide optimal protection.

 

IF BIRD FLU BREAKS OUT, WE HAVE A VACCINE

In the event of an outbreak of avian influenza in humans, the US government now has a vaccine against H5N1, the causative virus. A two-dose regimen of a whole-virus H5N1 vaccine, which is derived from cell culture, induced neutralizing antibodies against diverse H5N1 virus strains in most subjects in one study.³⁴ Another vaccine, which is egg-independent and adenoviral vector-based and contains conserved nucleoproteins, is broadly protective against globally dispersed H5N1 virus clades.³⁵ The addition of the MF59 adjuvant to a subvirion H5N1 vaccine increased antibody response, but the addition of aluminum hydroxide did not.³⁶

EXERCISE AND HYGIENE PREVENT FLU

Exercise has benefits beyond the usual ones: one study showed that exercising at low to moderate frequency (between once a month and three times a week) is associated with lower rates of influenza-related death.³⁷

A recent meta-analysis³⁸ confirmed that hygienic measures can prevent the spread of respiratory viruses in the community. The investigators calculated that hand-washing at least 10 times daily can prevent a large number of these infections (number needed to treat [NNT] = 4), and wearing surgical masks (NNT = 6), N95 masks (NNT = 3), gloves (NNT = 6), and gowns (NNT = 5) had incremental effects. On the other hand, the value of adding virucidal or antiseptic solutions to normal hand-washing was uncertain. Strict adherence to hand hygiene and masks (including by children) is needed to prevent influenza transmission in the home.³⁹

AMANTADINE, RIMANTADINE ARE OUT; OSELTAMIVIR RESISTANCE IS GROWING

The CDC continues to recommend against using amantadine (Symmetrel) or rimantadine (Flumadine) to treat flu, owing to a high rate (> 90%) of resistance to these drugs.

A nonrandomized study suggests that zanamivir (Relenza) is more effective than oseltamivir (Tamiflu) for treating influenza B.⁴⁰ A retrospective study in nine lung transplant recipients showed that oseltamivir is well tolerated and may reduce the risk of complications in these patients.⁴¹ Large, randomized, multicenter studies are under way to better assess oseltamivir’s preventive and therapeutic efficacy in transplant recipients.

In children, as in adults, oseltamivir is less effective against influenza B than influenza A,⁴² and both neuraminidase inhibitors, ie, oseltamivir and zanamivir, are equally effective in reducing the febrile period of influenza.⁴³

During the 2007–2008 season, the rate of resistance to oseltamivir increased alarmingly.⁴⁴ Resistance was restricted to A (H1N1) viruses carrying the H274Y mutation. In March 2008, the frequency of resistance among A (H1N1) viruses in the United States was 8.6%, 10 times higher than during the preceding influenza season. Resistance rates were much higher in several European countries, including Norway and France. During the Southern Hemisphere’s influenza season (May 2008 though September 2008), 46.5% of influenza A (H1N1) viruses received from 14 countries were resistant to oseltamivir. ⁴⁵ It is worrisome that many of these resistant viruses were isolated from untreated patients. Fortunately, to date, 99% of these isolates remain susceptible to zanamivir.

Microbiologic tests to detect resistance are not currently available for clinical use. During an influenza pandemic, widespread use of neuraminidase inhibitors will likely promote further development of drug resistance. A mathematical model concluded that combined treatment and prophylaxis with antiviral agents will be necessary to control transmission during a pandemic, and that allocating different drugs to cases and contacts would be most effective in curtailing emergence of resistance.⁴⁶

For now, either oseltamivir or zanamivir is acceptable for patients with flu symptoms and can be started pending results of PCR testing of nasopharyngeal swabs to make sure that the patient really has influenza. The drugs should be taken for 5 days.

Last year, some people may have lost their faith in flu shots. The three antigens chosen for the vaccine in advance by the US Centers for Disease Control and Prevention (CDC) did not match very well the influenza strains that ultimately circulated in North America, and the overall protective efficacy of the vaccine was estimated at only 40%.

Nevertheless, vaccination remains the primary preventive measure for both epidemic and pandemic influenza, especially in view of a rising rate of resistance to the oral antiviral agent oseltamivir (Tamiflu).

In the 2008–2009 influenza season, we hope to do better. All three antigens contained in the 2008–2009 vaccine are new. Surveillance data from the Southern Hemisphere during the summer of 2008 show that this vaccine is expected to be a good match for the strains circulating in the Northern Hemisphere. And with 146 million doses expected to be manufactured this season by six companies—the largest number of doses ever manufactured in the United States—enough should be available for all.

GREAT STRIDES HAVE BEEN MADE, BUT FLU IS STILL A PROBLEM

We are making great strides against influenza. Over the last 50 years, the rate of influenzarelated deaths in the United States declined by 95%, from an average seasonal rate of 10.2 deaths per 100,000 population in the 1940s to 0.56 per 100,000 by the 1990s.¹

However, influenza still accounts for about 10% of patients admitted to intensive care units for acute respiratory failure during epidemics.²

Children and the elderly are still hit the hardest: infants age 0 through 23 months and adults age 65 years and older have the highest peak rates of pneumonia and influenza hospitalization and death.³ School-age children (5–18 years) have an indirect role in anticipating the risk to others and can learn to help avoid spreading the virus by washing their hands more, wearing masks, and adopting other hygienic measures.

In the 1918–1919 pandemic, most deaths were from secondary bacterial pneumonia, a fact that has implications for pandemic preparedness. ⁴ Currently, Staphylococcus aureus, particularly methicillin-resistant strains (MRSA), is an important cause of secondary bacterial pneumonia, with a high mortality rate.⁵

UPDATE ON DIAGNOSIS: PCR IS THE BEST TEST

In the hospital, it is important to identify patients who have influenza so that we can give them appropriate antiviral therapy and also protect other patients from getting the flu. Unfortunately, the sensitivity and positive predictive value of fever, cough, and other symptoms for the diagnosis of influenza in hospitalized patients are 40% or less.⁶

Real-time reverse transcriptase polymerase chain reaction (PCR), compared with direct fluorescent antigen detection or cell culture, has the highest sensitivity (98.7%) and specificity (100%) in both children⁷ and the elderly.⁸ Furthermore, cell culture is slow and therefore is not useful in clinical practice. Nasopharyngeal wash sampling appears impractical in nursing home residents, owing to their underlying disabilities, and nasopharyngeal swabs tested by PCR are equally sensitive. ⁸

However, improvements are needed in molecular detection and subtyping of influenza viruses.⁹ If a pandemic breaks out, we will need to identify the virus quickly to have enough time for preventive interventions. The US Food and Drug Administration has recently cleared a new test called the Human Virus Real-Time RT-PCR Detection and Characterization Panel to detect and differentiate between seasonal and novel influenza strains.¹⁰

UPDATE ON INFLUENZA VACCINE

New recommendations in 2008 by the CDC Advisory Committee on Immunization Practices¹¹ include annual vaccination for all children age 5 through 18 years, and either the trivalent inactivated vaccine (ie, the shot) or the live-attenuated vaccine (ie, the Flu-Mist intranasal spray) for healthy people age 2 through 49 years. The CDC recommendations are summarized at www.cdc.gov/flu/professionals/acip/index.htm.

Has the benefit of vaccination in adults been overestimated?

Jackson et al,¹² in an article published in August 2008, suggested that the effect of influenza vaccination on the risk of community-acquired pneumonia in immunocompetent elderly people during influenza season is less than previously estimated. However, some patients in this study who were classified as not having been vaccinated may have actually been vaccinated by other health care providers without notifying their primary care providers. Moreover, influenza infection may cause only a small proportion of cases of pneumonia in this population.

In another study, Eurich et al¹³ suggested that previous observational studies overestimated the benefit of influenza vaccination on reducing deaths in patients with pneumonia outside the flu season. Although they found the incidence of death to be 51% lower in vaccinated than in unvaccinated adults with community-acquired pneumonia (N = 1,813) admitted to six hospitals, they ascribed it to confounding factors, specifically socioeconomic and functional status. This phenomenon was previously called the “frailty bias” or the “healthy user effect.” However, this study included only patients hospitalized with pneumonia and did not include data on vaccine-induced immunity or the cause of pneumonia, and measures of the healthy user effect were rudimentary. In addition, only outcomes during hospitalization were included.

Most experts still believe that vaccination prevents 50% of influenza-related deaths (with a smaller effect on rates of all-cause mortality¹⁴), including deaths in very old people. ¹⁵ A recent review found no basis for the historic concern that the antibody response to the influenza vaccine in people age 60 and older declines more rapidly than in younger people and below seroprotective levels within 4 months of immunization.¹⁶

Nevertheless, discordance between antibody and T-cell responses to influenza vaccine does exist¹⁷ (ie, the vaccine can induce antibodies while not boosting the T-cell-specific response), and we should continue to seek new vaccines that are more effective.

 

 

More people are being vaccinated, but we’re still below our goals

Although influenza vaccination rates among adults continue to improve,¹⁸ they remain well below the Healthy People 2010 initiative’s target of 90% in adults age 65 and older (the current rate is 72%) and below the target of 60% in people age 18 through 64 who have one or more high-risk conditions, health care workers, and pregnant women¹⁹ (currently 35% in people age 18 through 49 and 42% in people age 50 through 64). Thus, we still need to improve vaccination coverage rates.

Health care providers should offer vaccination at every opportunity between October and May.²⁰ Offering vaccination in nontraditional settings such as work sites and pharmacies is likely to be cost-saving for healthy adults due to averted morbidity.²¹ At many hospitals, health care workers can opt out of being vaccinated, but they must formally state that they are doing so. The use of these declination statements among health care workers is associated with a mean increase of 11.6% in vaccination rates.²²

Since influenza is the second most frequent vaccine-preventable infection in travelers, the vaccine should be offered to those crossing to the opposite hemisphere during its peak influenza season (eg, to South America in May through September), as well as to those visiting the tropics at any time of year.²³

Vaccination is safe and effective in high-risk groups

Data on vaccination are reassuring in several at-risk groups.

In pregnancy, there is no indication that infants are harmed if their mothers are vaccinated in the first trimester.²⁴ The evidence of excess morbidity during influenza epidemics supports vaccinating healthy pregnant women in the second or third trimester and those with comorbidities any time during pregnancy. Influenza vaccination during pregnancy reduces laboratory-confirmed influenza in infants up to 6 months of age by 63% and prevents 29% of all febrile respiratory illnesses in infants and 36% of those in mothers.²⁵

In patients with chronic obstructive pulmonary disease, vaccination cuts the rate of outpatient visits and hospitalizations due to acute respiratory illness by 67%.²⁶ The antibody response to influenza vaccine in patients with rheumatoid arthritis treated with rituximab (Rituxan), a monoclonal antibody directed against CD20 surface antigen-positive B lymphocytes, was lower than in healthy controls, but was not negligible.²⁷

Dispelling myths about vaccination in children

One recently published study in children younger than 5 years did not find vaccination to be effective in preventing influenza-related hospitalizations and outpatient visits.²⁸ However, in both seasons in which this study was conducted, there was a suboptimal antigenic match between vaccines and circulating strains. Moreover, about 60% of participants were unvaccinated and another 20% were only partially vaccinated, making it difficult to assess vaccine effectiveness. Several other studies have shown that, when there is a good match, vaccine effectiveness in children is 85% to 90%.

Even though the live-attenuated (inhaled) vaccine is more expensive than the inactivated (injected) vaccine, it reduces the number of influenza illness cases and lowers subsequent health care use in children and productivity loss in their parents, with a net total savings of $45.80 relative to the inactivated vaccine.²⁹ The live-attenuated vaccine provides sustained protection against influenza illness for 12 months following vaccination, as well as meaningful efficacy through a second season without revaccination, although at a lower level.³⁰

Several myths about the live-attenuated vaccine should be dispelled.³¹ It is well tolerated and causes only mild, transient symptoms of upper respiratory infection, even in people with asthma or the early stages of human immunodeficiency virus infection. Genetic reversion of the vaccine strain to a wild-type virus requires independent mutation in four gene segments, an event that has not been observed. Finally, although viral shedding is common for several days after vaccination, transmission to another person has been shown in only one person, who remained asymptomatic.

Unfortunately, rates of influenza vaccination are even worse for children than for adults.³² In children 6 through 23 months old, only 22% are fully vaccinated; in those 24 through 59 months old, only 16.5% are.

One group of immunocompromised children, liver transplant recipients, achieved antibody seroprotection and seroconversion rates similar to those achieved by their healthy siblings, with no vaccine-related serious side effects.³³ As in adults, the cell-mediated immune response to the vaccine was diminished, suggesting that other strategies are needed to provide optimal protection.

 

IF BIRD FLU BREAKS OUT, WE HAVE A VACCINE

In the event of an outbreak of avian influenza in humans, the US government now has a vaccine against H5N1, the causative virus. A two-dose regimen of a whole-virus H5N1 vaccine, which is derived from cell culture, induced neutralizing antibodies against diverse H5N1 virus strains in most subjects in one study.³⁴ Another vaccine, which is egg-independent and adenoviral vector-based and contains conserved nucleoproteins, is broadly protective against globally dispersed H5N1 virus clades.³⁵ The addition of the MF59 adjuvant to a subvirion H5N1 vaccine increased antibody response, but the addition of aluminum hydroxide did not.³⁶

EXERCISE AND HYGIENE PREVENT FLU

Exercise has benefits beyond the usual ones: one study showed that exercising at low to moderate frequency (between once a month and three times a week) is associated with lower rates of influenza-related death.³⁷

A recent meta-analysis³⁸ confirmed that hygienic measures can prevent the spread of respiratory viruses in the community. The investigators calculated that hand-washing at least 10 times daily can prevent a large number of these infections (number needed to treat [NNT] = 4), and wearing surgical masks (NNT = 6), N95 masks (NNT = 3), gloves (NNT = 6), and gowns (NNT = 5) had incremental effects. On the other hand, the value of adding virucidal or antiseptic solutions to normal hand-washing was uncertain. Strict adherence to hand hygiene and masks (including by children) is needed to prevent influenza transmission in the home.³⁹

AMANTADINE, RIMANTADINE ARE OUT; OSELTAMIVIR RESISTANCE IS GROWING

The CDC continues to recommend against using amantadine (Symmetrel) or rimantadine (Flumadine) to treat flu, owing to a high rate (> 90%) of resistance to these drugs.

A nonrandomized study suggests that zanamivir (Relenza) is more effective than oseltamivir (Tamiflu) for treating influenza B.⁴⁰ A retrospective study in nine lung transplant recipients showed that oseltamivir is well tolerated and may reduce the risk of complications in these patients.⁴¹ Large, randomized, multicenter studies are under way to better assess oseltamivir’s preventive and therapeutic efficacy in transplant recipients.

In children, as in adults, oseltamivir is less effective against influenza B than influenza A,⁴² and both neuraminidase inhibitors, ie, oseltamivir and zanamivir, are equally effective in reducing the febrile period of influenza.⁴³

During the 2007–2008 season, the rate of resistance to oseltamivir increased alarmingly.⁴⁴ Resistance was restricted to A (H1N1) viruses carrying the H274Y mutation. In March 2008, the frequency of resistance among A (H1N1) viruses in the United States was 8.6%, 10 times higher than during the preceding influenza season. Resistance rates were much higher in several European countries, including Norway and France. During the Southern Hemisphere’s influenza season (May 2008 though September 2008), 46.5% of influenza A (H1N1) viruses received from 14 countries were resistant to oseltamivir. ⁴⁵ It is worrisome that many of these resistant viruses were isolated from untreated patients. Fortunately, to date, 99% of these isolates remain susceptible to zanamivir.

Microbiologic tests to detect resistance are not currently available for clinical use. During an influenza pandemic, widespread use of neuraminidase inhibitors will likely promote further development of drug resistance. A mathematical model concluded that combined treatment and prophylaxis with antiviral agents will be necessary to control transmission during a pandemic, and that allocating different drugs to cases and contacts would be most effective in curtailing emergence of resistance.⁴⁶

For now, either oseltamivir or zanamivir is acceptable for patients with flu symptoms and can be started pending results of PCR testing of nasopharyngeal swabs to make sure that the patient really has influenza. The drugs should be taken for 5 days.

Last year, some people may have lost their faith in flu shots. The three antigens chosen for the vaccine in advance by the US Centers for Disease Control and Prevention (CDC) did not match very well the influenza strains that ultimately circulated in North America, and the overall protective efficacy of the vaccine was estimated at only 40%.

Nevertheless, vaccination remains the primary preventive measure for both epidemic and pandemic influenza, especially in view of a rising rate of resistance to the oral antiviral agent oseltamivir (Tamiflu).

In the 2008–2009 influenza season, we hope to do better. All three antigens contained in the 2008–2009 vaccine are new. Surveillance data from the Southern Hemisphere during the summer of 2008 show that this vaccine is expected to be a good match for the strains circulating in the Northern Hemisphere. And with 146 million doses expected to be manufactured this season by six companies—the largest number of doses ever manufactured in the United States—enough should be available for all.

GREAT STRIDES HAVE BEEN MADE, BUT FLU IS STILL A PROBLEM

We are making great strides against influenza. Over the last 50 years, the rate of influenzarelated deaths in the United States declined by 95%, from an average seasonal rate of 10.2 deaths per 100,000 population in the 1940s to 0.56 per 100,000 by the 1990s.¹

However, influenza still accounts for about 10% of patients admitted to intensive care units for acute respiratory failure during epidemics.²

Children and the elderly are still hit the hardest: infants age 0 through 23 months and adults age 65 years and older have the highest peak rates of pneumonia and influenza hospitalization and death.³ School-age children (5–18 years) have an indirect role in anticipating the risk to others and can learn to help avoid spreading the virus by washing their hands more, wearing masks, and adopting other hygienic measures.

In the 1918–1919 pandemic, most deaths were from secondary bacterial pneumonia, a fact that has implications for pandemic preparedness. ⁴ Currently, Staphylococcus aureus, particularly methicillin-resistant strains (MRSA), is an important cause of secondary bacterial pneumonia, with a high mortality rate.⁵

UPDATE ON DIAGNOSIS: PCR IS THE BEST TEST

In the hospital, it is important to identify patients who have influenza so that we can give them appropriate antiviral therapy and also protect other patients from getting the flu. Unfortunately, the sensitivity and positive predictive value of fever, cough, and other symptoms for the diagnosis of influenza in hospitalized patients are 40% or less.⁶

Real-time reverse transcriptase polymerase chain reaction (PCR), compared with direct fluorescent antigen detection or cell culture, has the highest sensitivity (98.7%) and specificity (100%) in both children⁷ and the elderly.⁸ Furthermore, cell culture is slow and therefore is not useful in clinical practice. Nasopharyngeal wash sampling appears impractical in nursing home residents, owing to their underlying disabilities, and nasopharyngeal swabs tested by PCR are equally sensitive. ⁸

However, improvements are needed in molecular detection and subtyping of influenza viruses.⁹ If a pandemic breaks out, we will need to identify the virus quickly to have enough time for preventive interventions. The US Food and Drug Administration has recently cleared a new test called the Human Virus Real-Time RT-PCR Detection and Characterization Panel to detect and differentiate between seasonal and novel influenza strains.¹⁰

UPDATE ON INFLUENZA VACCINE

New recommendations in 2008 by the CDC Advisory Committee on Immunization Practices¹¹ include annual vaccination for all children age 5 through 18 years, and either the trivalent inactivated vaccine (ie, the shot) or the live-attenuated vaccine (ie, the Flu-Mist intranasal spray) for healthy people age 2 through 49 years. The CDC recommendations are summarized at www.cdc.gov/flu/professionals/acip/index.htm.

Has the benefit of vaccination in adults been overestimated?

Jackson et al,¹² in an article published in August 2008, suggested that the effect of influenza vaccination on the risk of community-acquired pneumonia in immunocompetent elderly people during influenza season is less than previously estimated. However, some patients in this study who were classified as not having been vaccinated may have actually been vaccinated by other health care providers without notifying their primary care providers. Moreover, influenza infection may cause only a small proportion of cases of pneumonia in this population.

In another study, Eurich et al¹³ suggested that previous observational studies overestimated the benefit of influenza vaccination on reducing deaths in patients with pneumonia outside the flu season. Although they found the incidence of death to be 51% lower in vaccinated than in unvaccinated adults with community-acquired pneumonia (N = 1,813) admitted to six hospitals, they ascribed it to confounding factors, specifically socioeconomic and functional status. This phenomenon was previously called the “frailty bias” or the “healthy user effect.” However, this study included only patients hospitalized with pneumonia and did not include data on vaccine-induced immunity or the cause of pneumonia, and measures of the healthy user effect were rudimentary. In addition, only outcomes during hospitalization were included.

Most experts still believe that vaccination prevents 50% of influenza-related deaths (with a smaller effect on rates of all-cause mortality¹⁴), including deaths in very old people. ¹⁵ A recent review found no basis for the historic concern that the antibody response to the influenza vaccine in people age 60 and older declines more rapidly than in younger people and below seroprotective levels within 4 months of immunization.¹⁶

Nevertheless, discordance between antibody and T-cell responses to influenza vaccine does exist¹⁷ (ie, the vaccine can induce antibodies while not boosting the T-cell-specific response), and we should continue to seek new vaccines that are more effective.

 

 

More people are being vaccinated, but we’re still below our goals

Although influenza vaccination rates among adults continue to improve,¹⁸ they remain well below the Healthy People 2010 initiative’s target of 90% in adults age 65 and older (the current rate is 72%) and below the target of 60% in people age 18 through 64 who have one or more high-risk conditions, health care workers, and pregnant women¹⁹ (currently 35% in people age 18 through 49 and 42% in people age 50 through 64). Thus, we still need to improve vaccination coverage rates.

Health care providers should offer vaccination at every opportunity between October and May.²⁰ Offering vaccination in nontraditional settings such as work sites and pharmacies is likely to be cost-saving for healthy adults due to averted morbidity.²¹ At many hospitals, health care workers can opt out of being vaccinated, but they must formally state that they are doing so. The use of these declination statements among health care workers is associated with a mean increase of 11.6% in vaccination rates.²²

Since influenza is the second most frequent vaccine-preventable infection in travelers, the vaccine should be offered to those crossing to the opposite hemisphere during its peak influenza season (eg, to South America in May through September), as well as to those visiting the tropics at any time of year.²³

Vaccination is safe and effective in high-risk groups

Data on vaccination are reassuring in several at-risk groups.

In pregnancy, there is no indication that infants are harmed if their mothers are vaccinated in the first trimester.²⁴ The evidence of excess morbidity during influenza epidemics supports vaccinating healthy pregnant women in the second or third trimester and those with comorbidities any time during pregnancy. Influenza vaccination during pregnancy reduces laboratory-confirmed influenza in infants up to 6 months of age by 63% and prevents 29% of all febrile respiratory illnesses in infants and 36% of those in mothers.²⁵

In patients with chronic obstructive pulmonary disease, vaccination cuts the rate of outpatient visits and hospitalizations due to acute respiratory illness by 67%.²⁶ The antibody response to influenza vaccine in patients with rheumatoid arthritis treated with rituximab (Rituxan), a monoclonal antibody directed against CD20 surface antigen-positive B lymphocytes, was lower than in healthy controls, but was not negligible.²⁷

Dispelling myths about vaccination in children

One recently published study in children younger than 5 years did not find vaccination to be effective in preventing influenza-related hospitalizations and outpatient visits.²⁸ However, in both seasons in which this study was conducted, there was a suboptimal antigenic match between vaccines and circulating strains. Moreover, about 60% of participants were unvaccinated and another 20% were only partially vaccinated, making it difficult to assess vaccine effectiveness. Several other studies have shown that, when there is a good match, vaccine effectiveness in children is 85% to 90%.

Even though the live-attenuated (inhaled) vaccine is more expensive than the inactivated (injected) vaccine, it reduces the number of influenza illness cases and lowers subsequent health care use in children and productivity loss in their parents, with a net total savings of $45.80 relative to the inactivated vaccine.²⁹ The live-attenuated vaccine provides sustained protection against influenza illness for 12 months following vaccination, as well as meaningful efficacy through a second season without revaccination, although at a lower level.³⁰

Several myths about the live-attenuated vaccine should be dispelled.³¹ It is well tolerated and causes only mild, transient symptoms of upper respiratory infection, even in people with asthma or the early stages of human immunodeficiency virus infection. Genetic reversion of the vaccine strain to a wild-type virus requires independent mutation in four gene segments, an event that has not been observed. Finally, although viral shedding is common for several days after vaccination, transmission to another person has been shown in only one person, who remained asymptomatic.

Unfortunately, rates of influenza vaccination are even worse for children than for adults.³² In children 6 through 23 months old, only 22% are fully vaccinated; in those 24 through 59 months old, only 16.5% are.

One group of immunocompromised children, liver transplant recipients, achieved antibody seroprotection and seroconversion rates similar to those achieved by their healthy siblings, with no vaccine-related serious side effects.³³ As in adults, the cell-mediated immune response to the vaccine was diminished, suggesting that other strategies are needed to provide optimal protection.

 

IF BIRD FLU BREAKS OUT, WE HAVE A VACCINE

In the event of an outbreak of avian influenza in humans, the US government now has a vaccine against H5N1, the causative virus. A two-dose regimen of a whole-virus H5N1 vaccine, which is derived from cell culture, induced neutralizing antibodies against diverse H5N1 virus strains in most subjects in one study.³⁴ Another vaccine, which is egg-independent and adenoviral vector-based and contains conserved nucleoproteins, is broadly protective against globally dispersed H5N1 virus clades.³⁵ The addition of the MF59 adjuvant to a subvirion H5N1 vaccine increased antibody response, but the addition of aluminum hydroxide did not.³⁶

EXERCISE AND HYGIENE PREVENT FLU

Exercise has benefits beyond the usual ones: one study showed that exercising at low to moderate frequency (between once a month and three times a week) is associated with lower rates of influenza-related death.³⁷

A recent meta-analysis³⁸ confirmed that hygienic measures can prevent the spread of respiratory viruses in the community. The investigators calculated that hand-washing at least 10 times daily can prevent a large number of these infections (number needed to treat [NNT] = 4), and wearing surgical masks (NNT = 6), N95 masks (NNT = 3), gloves (NNT = 6), and gowns (NNT = 5) had incremental effects. On the other hand, the value of adding virucidal or antiseptic solutions to normal hand-washing was uncertain. Strict adherence to hand hygiene and masks (including by children) is needed to prevent influenza transmission in the home.³⁹

AMANTADINE, RIMANTADINE ARE OUT; OSELTAMIVIR RESISTANCE IS GROWING

The CDC continues to recommend against using amantadine (Symmetrel) or rimantadine (Flumadine) to treat flu, owing to a high rate (> 90%) of resistance to these drugs.

A nonrandomized study suggests that zanamivir (Relenza) is more effective than oseltamivir (Tamiflu) for treating influenza B.⁴⁰ A retrospective study in nine lung transplant recipients showed that oseltamivir is well tolerated and may reduce the risk of complications in these patients.⁴¹ Large, randomized, multicenter studies are under way to better assess oseltamivir’s preventive and therapeutic efficacy in transplant recipients.

In children, as in adults, oseltamivir is less effective against influenza B than influenza A,⁴² and both neuraminidase inhibitors, ie, oseltamivir and zanamivir, are equally effective in reducing the febrile period of influenza.⁴³

During the 2007–2008 season, the rate of resistance to oseltamivir increased alarmingly.⁴⁴ Resistance was restricted to A (H1N1) viruses carrying the H274Y mutation. In March 2008, the frequency of resistance among A (H1N1) viruses in the United States was 8.6%, 10 times higher than during the preceding influenza season. Resistance rates were much higher in several European countries, including Norway and France. During the Southern Hemisphere’s influenza season (May 2008 though September 2008), 46.5% of influenza A (H1N1) viruses received from 14 countries were resistant to oseltamivir. ⁴⁵ It is worrisome that many of these resistant viruses were isolated from untreated patients. Fortunately, to date, 99% of these isolates remain susceptible to zanamivir.

Microbiologic tests to detect resistance are not currently available for clinical use. During an influenza pandemic, widespread use of neuraminidase inhibitors will likely promote further development of drug resistance. A mathematical model concluded that combined treatment and prophylaxis with antiviral agents will be necessary to control transmission during a pandemic, and that allocating different drugs to cases and contacts would be most effective in curtailing emergence of resistance.⁴⁶

For now, either oseltamivir or zanamivir is acceptable for patients with flu symptoms and can be started pending results of PCR testing of nasopharyngeal swabs to make sure that the patient really has influenza. The drugs should be taken for 5 days.