Shingles linked to stroke risk

Friday October 9 2009

People with shingles are almost a third more likely to have a stroke in the next year, according to The Daily Telegraph, which reported research into stroke risk following shingles. The newspaper also says that people with a “rare” form of shingles affecting the eye were four times more likely to suffer from a stroke.

The research behind this report followed groups of people with and without shingles, a condition caused by the varicella virus (chickenpox) earlier in life being reactivated years later. This research cannot prove whether or not shingles causes a stroke because it was a cohort study, which cannot establish cause and effect.

The researchers were also unable to take into account factors including general health, obesity, smoking and physical activity, as these were not included in their original data source.

A simple explanation for the results could be that both shingles and stroke are more common in people with poorer health and unhealthier lifestyles.

There may be a link between adult shingles and stroke, but more robust research is needed to confirm the strength and nature of that link. There is not yet enough evidence to justify changing the way stroke is prevented or treated.

Where did the story come from?

The study was carried out by Dr Jiunn-Horng Kang and colleagues from the Taipei Medical University and the Taipei Medical University Hospital. The study was published in the peer-reviewed medical journal_ Stroke,_ which did not list sources of funding.

What kind of scientific study was this?

This was a retrospective cohort study examining the risk and frequency of stroke following herpes zoster (shingles) attacks.

The varicella zoster virus (VZV) causes chickenpox in children. Following infection, which is usually mild, the virus lies dormant in the nerve ganglia (groups of nerve cells outside the brain). In the elderly and people with weakened immune systems, reactivation of the infection leads to herpes zoster (shingles), which manifests itself as a rash and painful blisters on the skin along an infected nerve, usually on only one side of the body.

The researchers say that some previous studies have reported an association between herpes zoster and vascular problems and stroke.

In this study, researchers enrolled 7,760 people who had been treated for herpes zoster (shingles) between 1997 and 2001 and matched them with 23,280 randomly selected subjects that had not had herpes zoster treatment. At the end of a one-year period they compared the stroke-free survival rate between the groups.

Data for this study was drawn from a random, representative sample of one million people selected from the Taiwanese national health insurance programme, which covers approximately 98% of the population. The database is managed by the Taiwan National Health Research Institute.

Adults who had received outpatient treatment for shingles from 1997 to 2001 and who had not had a stroke before their first visit were eligible for the shingles group. In total, 7,760 people were recruited. A second group of unaffected control subjects was matched to those with shingles on the basis of age and gender. The groups were to be selected in a ratio of three to one, so a total of 23,280 unaffected people were enrolled.

The patients were followed up for a year from the date of their first outpatient visit, and the development of any kind of stroke was recorded. The rate of stroke was compared between the herpes zoster and non-herpes zoster groups to estimate the association between shingles and strokes of any kind. The researchers also performed some separate analyses, estimating the risk of different types of stroke (ischemic and intracerebral or subarachnoid haemorrhaging), risk of stroke according to gender and risk of stroke in people with herpes zoster and eye-related complications.

What were the results of the study?

Even after matching subjects and controls according to their age, people who had herpes zoster were more likely to have additional conditions such as high blood pressure, diabetes, coronary heart disease, kidney disease, vascular disease and heart failure.

During the one-year follow-up period 439 people across both groups developed strokes: 133 from the herpes zoster group (representing 1.71% of their group) and 306 from the comparison group (1.31% of their group). When the researchers accounted for age, gender, income, other medical conditions, urbanisation and region of residence, the risk of stroke was increased by 1.3 times, but this was only just statistically significant (HR 1.31, 95% CI 1.00 to 1.60).

Ischemic stroke was 1.3 times more common in people who had herpes zoster (HR 1.31, 95% CI 1.07 to 1.65), and subarachnoid haemorrhage was 2.8 times more common (HR 2.79, 95% CI 1.69 to 4.61).

In people with herpes zoster ophthalmicus (eye complications), the risk of stroke was 4.28 times greater than that of people without herpes zoster infection (HR 4.28, 95% CI 2.01 to 9.01).

What interpretations did the researchers draw from these results?

The researchers say that the risk for stroke is increased after a herpes zoster attack. They add that although it is well-documented that vascular disease is a complication following a herpes zoster attack, which may induce a stroke, this “does not fully account for the unexpected high risk of stroke in these patients”.

What does the NHS Knowledge Service make of this study?

According to the authors, previous infection with the varicella virus (chickenpox) is an important risk factor for stroke in children. In this study they set out to estimate how frequent stroke was in adults following herpes zoster attacks (shingles). There are several points to consider in relation to these results and the study design:

  • This research shares a shortcoming with other cohort studies, which is not being able to adjust for all possible confounding factors that, in this case, may be related to both shingles infection and stroke outcome. While the authors considered the influence of several other illnesses in their analyses, they say that “the immunologic status and general health condition” were not measured for these patients, and therefore could not be taken into account. On the basis of this limitation, they say that the possibility that stroke development is associated with underlying conditions cannot be excluded, and that more study is needed.
  • The authors highlight some other limitations in their work, including possible classification errors in their database entries and the inability to adjust for potential confounding factors such as obesity, physical activity, smoking, diet, family history and alcohol use. All of these factors have been associated with stroke, although their relationship with the herpes zoster virus is not yet known.
  • The researchers have performed numerous different analyses on the data, increasing the likelihood that the findings that were reported in newspapers were simply chance results.
  • It may not be valid to extrapolate risk factors for stroke in children to adults, because children’s strokes are often different from those that affect adults.
  • The study was conducted in the Taiwanese population, and may not be relevant to those outside of Taiwan or the Taiwanese health system.

There may be a link between shingles in adults and stroke but more robust research is needed to confirm the strength and nature of the link.

Analysis by Bazian
Edited by NHS Website