Viral load blips in children from the CHIPS cohort

Polly Clayden, HIV i-Base

A poster from CHIPS evaluated the characteristics, predictors and consequences of transient increases in viral load “blips” in children on HAART in the UK and Ireland.

For this evaluation blips were defined as =1 viral load =50copies/mL between two undetectable viral loads <50 copies/mL, <280 days apart, during sustained viral suppression, without a change in treatment (excluding single drug substitutions for simplification/personal reasons and single drug interactions).

The investigators found of the 595 of the 1065 children enrolled in CHIPS to December 2005 initiating HAART naive, 347 (58%) achieved sustained viral load <50copies/mL on either first or second line treatment.

Of these, 78 experienced an overall total of 109 blips (between 1-4 per child) with median viral load 137copies/mL (IQR: 73-374 copies/mL); 17 blips were >1000copies/mL (max 39,838 copies/mL).

The investigators found blips were more common during second-line therapy (28/100 child-years (CY) [95%CI 16-38]) and following a previous blip (19/100CY [12-30]) compared to during first-line therapy without prior blips (10/100CY [8-13]).

Blipping occurred at all ages from 1-15 years but rates decreased with age at HAART initiation (IRR=0.94 [0.89-1.00] per year older, p=0.06), but were higher in children on PI (75% nelfinavir) regimens (IRR=1.62 [1.10-2.39], p=0.02), and in those remaining below detection for longer (IRR 1.21 [1.05-1.39] per extra year suppressed, p=0.009).

The children’s CD4 and CD8 counts were similar pre/post blip (median difference CD4 -33 cells/mm3 (IQR -230,304 cells/mm3), p=0.89 and CD8=10 cells/mm3 (-236,315 cells/mm3), p=0.51). 43% of detectable viral loads during periods of suppression were blips rather than virological failure.

The investigators found no evidence for a higher rate of virological failure in children who had ever blipped (adjusted HR: 0.74 [95% CI 0.46-1.17] p=0.20, for having blipped compared to never blipped. Additionally children who blipped in the last year (adjusted HR 0.93 [0.59-1.47], p=0.77) or blipped to >1000 copies/mL (adjusted HR 0.84 [0.35-2.06], p=71) were not at higher risk of subsequent failure.

They concluded that blips are fairly common in children occurring in 22% with otherwise undetectable viral load <50 copies/mL. Blipping is more common in children: starting HAART at younger age; on second line therapy; after a previous blip; on non-NNRTI containing regimen and in children with undetectable viral loads for more than one year.

They noted that this contrasted to at least one adult study that found no significant predictors of blipping. “But blips have little effect on CD4 and CD8 and do not increase rate of virological failure” they wrote.


Shingadia D, Lee KJ, Pillay D et al. Viral load blips in HIV-infected children from the CHIPS cohort: what do they mean? 8th International Congress on Drug Therapy in HIV Infection, 12-16 November 2006, Glasgow. Poster abstract P359.

Links to other websites are current at date of posting but not maintained.