Transient detectable viremia and the risk of viral rebound in patients from the Swiss HIV Cohort Study

BMC Infectious Diseases

Table 2 Estimates of associations between covariates and subsequent viral rebound in models fit to data from both first and subsequent suppression episodes

Covariate	Hazard ratio (95 % confidence interval)
	Model for interval censored data [24]	Gap-time Cox model [2]
Magnitude of first blip (reference no blips)
Low (50–199 copies/mL)	1.20 (0.89, 1.61)	1.03 (0.77, 1.38)
Medium (200–499 copies/mL)	1.42 (0.96, 2.19)	1.25 (0.86, 1.82)
High (500–999 copies/mL)	1.93 (1.24, 3.01)	1.69 (1.10, 2.60)
Female	1.16 (1.03, 1.30)	1.12 (1.00, 1.25)
Injection drug use^a	1.21 (1.04, 1.41)	1.25 (1.08, 1.45)
Age (per 10 years)	0.91 (0.86, 0.97)	0.89 (0.84, 0.95)
Calendar year episode began (per year)	0.92 (0.90, 0.94)	0.93 (0.91, 0.95)
Assay (reference Roche Amplicor ultrasensitive)
Abbot RealTime	2.19 (0.65, 7.41)	2.46 (0.81, 7.47)
Roche TaqMan version 1	0.90 (0.65, 1.25)	0.87 (0.63, 1.19)
Roche TaqMan version 2	1.31 (0.93, 1.86)	1.23 (0.88, 1.72)
cART regimen (reference NNRTI based)
Boosted PI	1.85 (1.62, 2.10)	1.76 (1.55, 2.01)
Single PI^b	1.73 (1.39, 2.16)	1.68 (1.35, 2.09
Entry or integrase inhibitor^c	1.98 (1.63, 2.41)	1.77 (1.45, 2.15)
None^d	10.6 (8.17, 13.7)	8.70 (6.87, 11.0)
CD4 cell count (per 100 cells/μL)^e
0 to <200	0.74 (0.57, 0.97)	na
≥ 200	1.01 (0.98, 1.03)	na
RNA tests per year (reference >6)^e
≤ 3	na	0.31 (0.23, 0.42)
3- ≤ 4	na	0.36 (0.27, 0.48)
4- ≤ 6	na	0.47 (0.27, 0.48)

cART combination antiretroviral therapy; NNRTI non-nucleoside reverse transcriptase inhibitor; PI protease inhibitor; na not applicable
^aInjection drug use as the most likely mode of HIV infection
^bAlso includes regimens with three nucleoside or nucleotide reverse transcriptase inhibitors - such regimens were considered cART if they followed another earlier cART regimen
^cAlso includes regimens with more than one PI (other than ritonavir), or with a PI and an NNRTI - all these regimens were mostly used as salvage regimens during this era
^dThe cART regimen was updated whenever its value changed within a suppression episodes. A patient not on cART was highly likely to experience viral rebound
^eThe gap-time Cox model in [2] has the number of RNA tests per year as a covariate but not CD4 cell count. The number of RNA tests per year is not an appropriate covariate in models for interval censored data – see, Additional file 1: Appendix A. Current (time updated) CD4 cell count was added to the model for interval censored data because it is a strong predictor of HIV progression even in patients with a suppressed viral load [27]

ISSN: 1471-2334