Thomas SL, Wheeler JG, Hall AJ
The Lancet. 2002;360(9334):678-682

 

It is still unknown whether the rates of developing herpes zoster are lower if individuals with latent varicella infection are exposed, or exposed often, to other individuals with either varicella or zoster. Many are concerned that universal immunization against varicella will result in more cases of zoster as the rate of wild-type varicella infection (and therefore the opportunity for close contacts of infected individuals to have their own immunity "boosted") declines. This study was a case-control design conducted in adult medicine practices in London. There were 244 subjects included during the 15-month study period, and 485 (2 per case) controls matched for age and gender within each practice. Cases were verified by PCR or by conservative clinical definitions (duration of pain, morphology and location of rash, etc). Controls were not included if they had any zoster-like rash in the 10 years preceding their interview. These were all immunocompetent individuals.

Estimates of exposure to cases of zoster or varicella were obtained by interviewing cases and controls. In addition to exposure to known varicella or zoster, the authors also assessed the degree to which the subjects were exposed to children -- socially or by occupation. The subjects were classified into 3 groups: exposed to known varicella or zoster, exposure to few children (household children, essentially), and exposure to "many" children ("many" was defined several ways, including occupations working with children).

In general, increasing contact with children, whether social or occupational, was associated with lower odds of developing zoster, and there appeared to be a dose-response effect. However, most of the associations were not statistically significant, although trend analyses were statistically significant (suggesting that larger numbers of subjects would have resulted in significant differences). There was a strong and statistically significant protective effect of known exposure to varicella. Again, there was a dose-response effect. An individual exposed to 1 known varicella case in the past 10 years had 0.90 odds of developing zoster compared to someone with no known varicella exposure. If an individual were exposed to 5 or more individuals with varicella, the odds of developing zoster was 29% of the risk of any nonexposed individual, even with adjustment for confounding variables. There was no protective effect of exposure to known cases of herpes zoster. The authors concluded that exposure to varicella is protective against developing zoster in healthy adults with latent varicella.

Reviewer comment: Although the subjects for this study were adults, the question evaluated in the study is on the minds of many pediatricians. It actually lends support to the concern that we will lose the population "boosting" effect of having wild-type varicella in our midst. It will be interesting to follow US surveillance data about zoster over the next few decades. Many pediatricians would not be surprised if some day down the road we are administering varicella booster immunizations to older individuals, and the authors note in their discussion section that such a trial is under way.

Ronkainen J, Nuutinen M, Koskimies O
The Lancet. 2002;360(9334):666-670

 

While pediatricians know that the biggest concern in children with Henoch-Schönlein purpura (HSP) is the potential for long-term renal involvement and injury, there is an ongoing need to obtain follow-up data on individuals afflicted as children. This study had a retrospective cohort design and re-evaluated 65 individuals diagnosed with HSP in Helsinki from 1964-1983. The mean age of the subjects at follow-up was 24 years. There were 29 individuals who had renal involvement at diagnosis, and 38 who did not. Fifty-two (80%) completed the survey to obtain follow-up medical history and 47 (72%) were examined by a study physician. Subjects were less likely to participate in the physician evaluation if they had no renal involvement at initial diagnosis.

Of the 47 individuals examined by a physician for this follow-up, 64% did not have renal disease. An additional 17% had "minor" abnormalities, including intermittent hypertension, proteinuria, or hematuria. Fifteen percent had active renal disease, and 4% had received a kidney transplant. Thirty-five percent of the individuals who had higher-grade nephritis at initial diagnosis or persistent nephritis after initial diagnosis had either active or end-stage renal disease at follow-up, compared with only 7% of children who had mild disease (hematuria or proteinuria, but no nephritis) or no renal involvement initially. The authors conclude that the absence of renal disease 1 month after HSP diagnosis is reassuring, but that all other children need long-term follow-up.

Reviewer comment: The take-home message here is that the presence of true nephritis at 1 month after diagnosis is predictive of who is at risk for ongoing problems. If there is renal disease beyond hematuria or proteinuria at 1 month, the children need even longer follow-up. It is also reassuring and helpful when counseling families to know that such a large proportion of even the individuals with significant renal involvement at diagnosis had good long-term outcomes.