Low body weight and type of protease inhibitor predict discontinuation and treatment-limiting adverse drug reactions among HIV-infected patients starting a protease inhibitor regimen: consistent results from a randomized trial and an observational cohort


  • O Kirk,

  • J Gerstoft,

  • C Pedersen,

  • H Nielsen,

  • N Obel,

  • Tl Katzenstein,

  • L Mathiesen,

  • Jd Lundgren

  • Correspondence: Ole Kirk, Department of Infectious Diseases, Hvidovre Hospital, University of Copenhagen, 2650 Hvidovre, Denmark. Tel: + 45 36 32 30 15; fax: + 45 36 47 33 40; e-mail: oki@cphiv.dk


Departments of Infectious Diseases,1Hvidovre Hospital, University of Copenhagen, Hvidovre,2Rigshospitalet, University of Copenhagen, Copenhagen,3Odense University Hospital, Odense,4Aalborg Hospital, Aalborg and5Marselisborg University Hospital, Aarhus, Denmark

Objectives  To assess predictors for discontinuation and treatment-limiting adverse drug reactions (TLADR) among patients starting their first protease inhibitor (PI).

Methods  Data on patients starting a PI regimen (indinavir, ritonavir, ritonavir/saquinavir and saquinavir hard gel) in a randomized trial (RAS, n = 318) and an observational cohort (OBC, n = 505) were used to document reasons for discontinuation and TLADR. Risk factors for discontinuation of the initial

PI/developing TLADR were assessed in Cox models.

Results  A total of 43 (RAS) and 48% (OBC) discontinued the initial PI therapy within less than 2 years. In both populations TLADR were the most common reason for discontinuation. The incidence of TLADR in RAS was: 8.5 (indinavir), 66.0 (ritonavir), 15.6 (saquinavir hard gel) per 100 person-years of follow-up (P < 0.001). Body weight and type of PI initiated were independent risk factors for treatment discontinuation and TLADR in both groups. In OBC, the risk of developing TLADR increased by 12% per 5 kg lower body weight when starting the PI regimen [the relative hazard (RH) was 1.12 (95% confidence interval: 1.05–1.19) per 5 kg lighter], and starting ritonavir was associated with a three- to sixfold higher risk of TLADR relative to other PI regimens. Very similar results were documented in RAS [RH for body weight was 1.18 (1.07–1.29)].

Conclusions  Nearly half of the patients stopped treatment with the initial PI, most commonly as a result of adverse drug reactions. Low body weight and initiation of ritonavir relative to other PIs were associated with an increased risk of TLADRs. Very consistent results were found in a randomized trial and an observational cohort.