Results of the NICHD clinical trial of the efficacy of intravenous immunoglobulin (IVIG) for the prophylaxis of serious bacterial infections in symptomatic HIV-infected children.

National Institute of Child Health and Human Development (NICHD)
January 18, 1991

Abstract:

NIH's National Institute of Child Health and Human Development (NICHD) has announced a treatment that reduces the development of serious bacterial infections in children infected with the human immunodeficiency virus (HIV). Bacterial infections in HIV-infected children are increased because of their impaired immune systems. The treatment, monthly administration of intravenous immunoglobulin (IVIG), also results in fewer hospitalizations for children with symptoms associated with HIV infection.

Between March 1, 1988, and Oct. 31, 1990, the NICHD enrolled 372 children between the ages of 2 months and 12 years who had experienced symptoms as a result of HIV infection into a clinical trial to test the effectiveness and safety of IVIG in preventing bacterial infections. The children were drawn from 28 clinical centers.

In IVIG treated children, treatment was shown to increase significantly the time free from serious bacterial infections, to reduce the number of overall bacterial infections, and to reduce the number of hospitalizations required in symptomatic HIV-infected children with CD4 lymphocyte counts of 200/mm3 or above. In children with CD4 counts below 200, no significant improvement was noted. CD4 lymphocyte measurements are one of the laboratory tests used to monitor the progression of HIV infection.

The clinical trial was scheduled to continue for another year but the results were so compelling that the study's Data Safety Monitoring Board recommended on Jan. 10 that the study be terminated. The investigators in the 28 centers have been notified of the demonstrated efficacy and safety of IVIG treatment.

Of the 372 infected children enrolled in the trial, all had at least some clinical or laboratory symptoms. They were categorized by CD4 lymphocyte count at the time they entered the study and by previous infection history. 115 had CD4 lymphocyte counts below 200 at the time they joined the study and/or a history of AIDS-defining opportunistic infections or recurrent serious bacterial infections. A second group of 257 children included those with lymphoctye counts above 200 and immunologic abnormalities or HIV-related clinical symptoms. 185 of the children received IVIG; 187 received a placebo (serum albumin). All patients in the study were permitted to receive the regular standard of care for HIV-infected children.

By the end of the study, nearly half of the children were receiving treatment to prevent Pneumocystis carinii pneumonia, a severe form of pneumonia often found in HIV positive individuals, and about 40 percent of the children were receiving AZT. The children receiving these additional treatments were almost equally divided among those receiving IVIG and those receiving placebo. These agents did not appear to affect the study's outcome.

During the study, 131 children experienced one or more serious laboratory-proven or clinically diagnosed bacterial infections, 77 in the placebo group compared to 54 in the IVIG group. The differences were significant in the group of children with higher lymphocyte counts (200/mm3 or above at entry). In that group, 57 of 151 children in the placebo group developed one or more serious laboratory-proven bacterial or clinically diagnosed infections compared to 38 of 158 children receiving IVIG. No differences in mortality were seen. There were 31 deaths in both the treatment and placebo groups.

In the group of children who entered the study with CD4 counts higher than 200 and who also had a prior history of AIDS-defining infections, IVIG treatment significantly prolonged bacterial infection-free time, 221 days for the placebo group compared to 583 days in the IVIG group. For children with CD4 counts less than 200, the infection-free duration was similar, 255 days in the placebo group compared to 282 days in the IVIG group.

Similarly, children with CD4 counts 200 or above and receiving IVIG had fewer overall bacterial infections when all infections were included. Children receiving placebo experienced 1.4 times more bacterial infections of any kind than children receiving IVIG (646 infections in placebo compared to 473 in IVIG). In addition, a reduction in the number of required hospitalizations was observed in children with high CD4 counts (200/mm3 or above) receiving IVIG. Overall, there were approximately 45 excess hospitalizations observed per 100-patient years in the placebo group compared to the IVIG group. There were few side effects noted among the children receiving IVIG; the side effects that were noted were mild, most often a brief rash. For additional information, call 1-800-TRIALS-A. To receive a four-page detailed summary of the clinical trials results by FAX or mail, please phone NLM's Office of Public Information at (301)-496-6308.

Full Text:

This advisory contains preliminary information for pediatric health care providers on the findings of the NICHD Clinical Trial of the Efficacy of IVIG in the Treatment of Symptomatic Children Infected with human immunodeficiency virus (HIV). While no recommendations for clinical practice are being made at the current time, the NIH wishes health-care providers to have additional information on this study while publication is being readied for submission to peer-reviewed medical journals.

Study Description

The study was a randomized, double-blind, controlled comparison of IVIG, 400 mg/kg infused every 28 days, versus albumin placebo for the prophylaxis of serious bacterial infections in immunologically or clinically symptomatic HIV-infected children less than 13 years of age. All study patients were permitted to receive the prevailing medical standard of care, including prophylaxis for Pneumocystis carinii pneumonia and Zidovudine (AZT) therapy.

Study patients were randomized to IVIG or albumin placebo within two groups on the basis of entry CD4 lymphocyte count and previous infection history. Group I included children with entry CD4 < 200/mm3 or a history of AIDS-defining opportunistic infections (CDC class P2D1) or recurrent serious bacterial infections (CDC class P2D2). Group II included children with entry CD4 >= 200/mm3 and CDC classification of P1B (immunologic abnormalities only), or other P2 categories (excluding those in Group I or children with malignancies, class P2E).

This trial enrolled 372 symptomatic HIV-infected children from 28 medical centers in the mainland United States and Puerto Rico between March 1, 1988 and October 31, 1990. Median follow-up was 17 months. Ninety-one percent of enrolled children had perinatally-acquired HIV infection; HIV-infected hemophiliac children were not included in the study population. The mean age of children at enrollment was 40 months, ranging from 2 months to 11 years; 78% of children were under the age of 5 years at entry. At entry, 12% of children had immunologic symptoms only (CDC class P1B), with the remainder of children being clinically symptomatic (CDC class P2). Prophylaxis for Pneumocystis carinii pneumonia (48% overall) and treatment with AZT (39% overall), was similar in both placebo and treatment groups. Overall mortality in the study group as a whole was 17%.

Primary study endpoints were the occurrence of laboratory-proven serious bacterial infection (meningitis, bacteremia, osteomyelitis, septic arthritis, acute sinusitis, pneumonia, acute mastoiditis, or abscess of an internal organ) or death. Secondary endpoints included clinically-diagnosed serious infections, other infections, and hospitalizations.

Results

Data from the study were reviewed January 10, 1991, by an independent Data Safety and Monitoring Board, which found that in those children with entry CD4 lymphocyte counts of 200/mm3 or greater, IVIG significantly prolonged the time free from serious laboratory-proven bacterial or clinically-diagnosed infections (24 month infection-free survival, 68% in IVIG group children compared to 48% in placebo group children, p = 0.005).

As of October 31, 1990, 48 children in the study experienced one or more laboratory-proven serious bacterial infections. Thirty children in the placebo group developed one or more laboratory-proven serious bacterial infections compared to 18 IVIG group patients. A similar reduction in serious infections with IVIG treatment was seen when clinically-diagnosed serious infections (pneumonia and acute sinusitis) were included.

In Group I children with entry CD4 >= 200/mm3 randomized to receive IVIG, the median time to development of a serious laboratory-proven bacterial or clinically-diagnosed infection was approximately 1 year longer than that observed for children randomized to the placebo arm (583 days in the IVIG children versus 221 days in the placebo children). Eighteen of 26 placebo group patients developed one or more serious laboratory-proven bacterial or clinically-diagnosed infections, compared to 14 of 34 children receiving IVIG (p=0.005 for 24 month infection-free survival). However, for Group I children entering with CD4 count less than 200/mm3, there was no difference in infection-free survival between the placebo and IVIG groups (282 days in the IVIG children versus 255 days in the placebo group).

In Group II, the proportion of children who developed serious laboratory-proven bacterial and clinically-diagnosed infections was less than observed in children in Group I. Forty-two children in the placebo group developed one or more serious laboratory-proven bacterial or clinically-diagnosed infections compared to 26 IVIG group children (p = 0.03 for 24 month infection-free survival). A 21% difference in infection-free survival at 24 months is observed between children receiving IVIG and those receiving placebo (75% 24-month infection-free survival in the IVIG group compared to 54% in the placebo group).

Mortality in the study group as a whole was 17% (median follow-up 17 months). There was no difference in mortality between the placebo and IVIG treatment groups, with 31 deaths in each arm.

Bacteremias accounted for the majority (75%) of laboratory-proven serious bacterial infections. The most common bacterial isolate was Streptococcus pneumoniae, accounting for 19 episodes of bacteremia. Fourteen episodes of pneumococcal bacteremia were noted in 12 children in the placebo group, compared to 5 episodes in 4 children receiving IVIG. The most striking effect of IVIG treatment on the incidence of pneumococcal bacteremia was noted in Group II children, where 10 episodes occurred in placebo patients, compared to 0 in children receiving IVIG.

Pneumonia accounted for the majority (65%) of clinically-diagnosed serious infections. All reported pneumonia and sinusitis episodes underwent independent review by two physicians blinded to the child's treatment status for classification as acute or not-acute episodes for study purposes. Children randomized to placebo were 1.5 times more likely to develop acute pneumonia than children receiving IVIG (66 episodes in 51 of 187 placebo group children compared to 43 episodes in 29 of 185 IVIG group children).

When all reported presumed bacterial infections, serious and minor and laboratory-proven as well as clinically-diagnosed, are included, a reduction in all infections is observed in IVIG group children. Children randomized to the placebo group experienced 1.4 times more presumed bacterial infections of any nature than did children who received IVIG (646 infections observed in placebo group children compared to 473 infections in IVIG group children, p = 0.02).

IVIG treatment was associated with a reduction in the number of hospitalizations (excluding social and short-term (one day or less) hospitalizations) for children with entry CD4 count >= 200/mm3. Overall, there were approximately 45 excess hospitalizations observed per year for every 100 HIV-infected children with CD4 count 200/mm3 or above in the placebo group when compared to the IVIG group.

Adverse reactions to study drug infusions were minimal and similar between albumin placebo and IVIG groups, noted in 0.6% of 3,203 placebo infusions and 0.5% of 3,199 IVIG infusions.

Conclusions

The results of this study indicate that IVIG was effective in prolonging infection-free time in immunologically or clinically symptomatic HIV-infected children with CD4 counts 200/mm3 or above. In children with CD4 counts < 200/mm3, efficacy was not demonstrated and other interventions may be needed to prevent serious bacterial infections. The study population did not include hemophiliac children, and while a significant proportion of children were five years old or older, 78% of children enrolled in this study were under 5 years of age, and the vast majority of children (91%) had perinatally-acquired HIV infection. Therefore, the possible efficacy of IVIG in hemophiliac children was not addressed by this study. Further evaluation of this and other ongoing studies may delineate further a subgroup of symptomatic HIV-infected children that would be most likely to benefit from the initiation of IVIG therapy.

This study supports the importance of identification of HIV-infected children to permit appropriate immunologic evaluation by their physician, and the initiation of appropriate adjunctive therapy (such as IVIG therapy and PCP prophylaxis) and, when appropriate, antiretroviral therapy.

For additional information, please call 1-800-TRIALS-A.