Flu + pneumococcal vaccines in elderly: compound benefits?

Oct 14, 2010 (CIDRAP News) – A large prospective study from Hong Kong raises the possibility that elderly people with chronic illnesses can significantly reduce their risk of stroke and heart attack by getting vaccinated against influenza and pneumococcal disease at the same time, but other researchers are greeting the findings with a mix of caution and skepticism.

The Hong Kong researchers recruited more than 36,000 elderly people with chronic illnesses to participate in the study, according to their recent report in Clinical Infectious Diseases. About 7,000 of them chose to receive both flu and 23-valent pneumococcal vaccines, while about 25,000 decided against the shots and lesser numbers picked one shot or the other.

After about a year of observation, those who received both immunizations had lower rates of death and of hospitalization for pneumonia, ischemic stroke, and myocardial infarction (MI) than those who received neither vaccine, with the reductions ranging from 33% to 48%. Smaller reductions in risk were seen in those who received just one of the shots.

"Dual vaccination with PPV [pneumococcal pneumonia vaccine] and TIV [trivalent influenza vaccine] is effective in protecting elderly persons with chronic illness from developing complications from respiratory, cardiovascular, and cerebrovascular diseases, thereby reducing hospitalization, coronary or intensive care admissions, and death," says the report by Ivan F. N. Hung and colleagues.

But others who have done research in the field say not so fast. They point out that another study a few months ago found no reduction in MI or stroke risk associated with pneumococcal vaccination. Further, they say the authors didn't assess whether outcomes like MI and stroke in the unvaccinated people were preceded by or associated with influenza or pneumococcal disease. Nor, they say, did the authors adequately address the possibility that the vaccinated people were generally healthier than the unvaccinated group to begin with—the "healthy vaccinee" effect.

Mass recruiting
All patients 65 years and older who had a chronic illness and attended outpatient clinics in the Hong Kong West Cluster from Dec 3, 2007, to Jun 30, 2008, were included in the study and were invited to receive the PPV and TIV. Chronic illnesses included asthma, chronic obstructive pulmonary disease (COPD), coronary artery disease, hypertension, diabetes, stroke, chronic renal or liver disease, or cancer.

The patients were invited to choose whether to receive both vaccines, just one, or none. The authors adjusted statistically for differences among the vaccination groups in a number of variables: age, sex, smoking history, and presence of the various chronic diseases. The patients were observed until Mar 31, 2009. The main outcome measure was the rate of death due to pneumonia, COPD, asthma, influenza-like illness (ILI), ischemic stroke, acute MI, or heart failure. Secondary outcome measures were hospital admissions due to the foregoing conditions and to pneumococcal pneumonia or ischemic heart disease.

Of 36,636 patients enrolled, 7,292 (19.9%) chose to receive both the PPV and TIV, 2,076 (5.7%) received TIV alone, 1,875 (5.1%) received PPV only, and 25,393 (69.3%) were unvaccinated. All the groups had similar levels of risk factors for poor outcomes, except that COPD was more common in the vaccinated groups than the unvaccinated.

Reductions in deaths and illnesses
At the end of the observation period, "dual vaccinees" had 35% fewer deaths from the relevant conditions than the unvaccinated group (hazard ratio [HR], 0.65, 95% confidence interval [CI], 0.55 to 0.77). Compared with the no-vaccine group, the dual vaccinees also had 43% fewer cases of pneumonia (HR, 0.57; 95% CI, 0.51 to 0.64), 33% fewer ischemic strokes (HR, 0.67; 95% CI, 0.54 to 0.83), and 48% fewer acute MIs (HR, 0.52; 95% CI, 0.83 to 0.71).

The authors also report significant reductions in hospitalizations for pneumococcal pneumonia, COPD, asthma, ILI, ischemic heart disease, and heart failure, compared with the unvaccinated group. In addition, they found that dual vaccination was associated with a 55% lower risk of admission to intensive care and a 41% lower risk of admission to a coronary care unit.

The researchers also found some other significant differences in outcomes between the various groups. For example, the dual-vaccine group had a 24% lower rate of pneumonia than the TIV-only group and a 26% lower rate of pneumonia than the PPV-only group. Also, the TIV-only group had fewer deaths, cases of pneumonia, COPD, and ILI than the unvaccinated group.

The researchers say theirs is the first study to carefully investigate the effects of dual vaccination on hospitalization for respiratory, cardiovascular, and cerebrovascular disease. Many previous studies of PPV or TIV alone, they say, did not specify whether patients might have received the other vaccine around the same time, potentially confounding the results.

They contend that the protective effect of dual vaccination is probably related to the prevention of acute infections and the resulting inflammatory responses, both systemic and local.

The authors acknowledge several limitations of their study. For ethical reasons, the participants were not randomly assigned, and the follow-up period was relatively short. They refer to the possibility of a "healthy vaccinee" effect, in that "health-conscious persons may be the ones who accept the vaccines, whereas non-health-conscious persons refuse." They were not able to analyze such possible confounding factors as the patients' diet and exercise habits.

"Despite these limitations, the conclusion drawn from this large prospective cohort study are highly valid as the comparison was made among participants from 4 groups of different vaccination status in a population of similar baseline characteristics, risk factors, socioeconomic status, [and] educational level," and because none of the patients had received PPV before, the authors conclude.

They note that after reviewing the findings, Hong Kong's government decided to provide free pneumococcal vaccination to elderly people with chronic illness, to supplement the free flu shots.

Views of others
Dr. Kristin Nichol, associate chief of staff at the Minneapolis Veterans Affairs Hospital and an experienced flu vaccination researcher, said she found the report "tantalizing" but not conclusive.

She said the findings must be viewed in the context of other research addressing similar questions. Some studies have failed to find a link between pneumococcal vaccination and a reduced risk of cardiovascular events, she said. In particular, a study published in the Journal of the American Medical Association in May found no reduction in the risk of MI or stroke in 45- to 69-year-old men, she explained.

"The Hong Kong study suggests that there is additional benefit with dual vaccination as opposed to flu vaccination alone," Nichol said. "I think it's tantalizing and interesting information, but not definitive, just because there are other studies that haven't demonstrated with pneumococcal vaccination a similar reduction in MI and stroke."

She added that there is substantial evidence that pneumococcal vaccination in the elderly reduces the risk of invasive pneumococcal disease with bacteremia, and that benefit clearly makes the immunization worthwhile. But studies have not convincingly demonstrated that pneumococcal vaccination reduces the overall risk of hospitalization, "and the question about MI and stroke with pneumococcal vaccination is less well-established," she said.

Nichol praised the Hong Kong authors for tackling such a large prospective study. "It does add to the literature," she said. "They did find a positive benefit with regard to MI and stroke. I'd say stay tuned to see if we can continue teasing out that question."

Edward Belongia, MD, director of the Epidemiology Research Center at the Marshfield Clinic Research Foundation in Marshfield, Wis., was more skeptical of the findings. He saw two main problems: the likelihood of a healthy vaccinee effect and the use of outcome measures not specifically related to influenza or pneumococcal disease.

"The huge reductions in many nonspecific outcomes (ischemic stroke, COPD, MI, etc) raise a big concern about unmeasured confounding," he commented. "We have already seen how the healthy vaccinee effect can lead to erroneous findings in observational studies of flu vaccine effectiveness that use similar nonspecific outcome measures, and the same type of confounding is very likely to be present in this study.

"The authors did not identify or adjust for measures of frailty in their analysis, and they were not able to show that any of the patients had laboratory-confirmed influenza or pneumococcal infection that preceded the outcomes that they studied. Without that, the biologic plausibility is highly questionable in my mind."

Though the groups in the study generally had the same prevalence of chronic conditions, that doesn't necessarily mean they had the same level of general health or frailty, because the chronic illnesses were identified by diagnostic codes, which don't address the severity of an illness, Belongia said. To get a more accurate picture would have required reviewing the patients' charts, he added. Thus, the dual-vaccination group may have had better initial health than the unvaccinated group.

Belongia stressed his view that the findings are limited by the lack of data on whether the various outcomes, such as hospitalizations for MI, stroke, heart failure, asthma, or COPD, were preceded by or associated with confirmed influenza or pneumococcal infections.

If a patient was healthy and then had an acute MI with no preceding illness, "it's a bit of a stretch to say the fact that they didn't get pneumococcal vaccine or TIV was a factor," he said.

The findings as stated "suggest we could prevent a large proportion of MIs just by giving TIV and pneumococcal vaccine," Belongia added. "That's an extraordinary magnitude of effect, and I think you need stronger evidence that does not rely on nonspecific outcomes."

"Our view, at least on influenza vaccine studies, is that these studies really need to have lab-confirmed infections," he said. "We need to stop doing studies with these nonspecific outcomes where there's a huge potential for confounding."

Hung IFN, Leung AYM, Chu DWS, et al. Prevention of acute myocardial infarction and stroke among elderly persons by dual pneumococcal and influenza vaccination: a prospective cohort study. Clin Infect Dis 2010 Nov 1;51 (early online publication) [Abstract]

See also:

Sep 21 CIDRAP News story "Study on flu shots and heart-attack risk questioned"

This week's top reads