Objective.-The clinical care of people infected with human immunodefic
iency virus (HIV) has been substantially affected by the introduction
of HIV-specific protease inhibitors (PIs). The 4 PIs available are saq
uinavir mesylate, ritonavir, indinavir sulfate, and nelfinavir mesylat
e. Comparison studies have not been reported; therefore, an assessment
of the available data to aid clinicians and patients in choosing appr
opriate treatment will be presented. Data Sources.-A systematic review
of peer-reviewed publications, abstracts from national and internatio
nal conferences, and product registration information through Septembe
r 1996. Study Selection and Data Extraction.-Criteria used to select s
tudies include their relevance to PIs, having been published in the En
glish language, and pertinence for clinicians. Data quality and validi
ty included the venue of the publication and relevance to clinical car
e. Data Synthesis.-Oral adminstration of ritonavir, indinavir, or nelf
inavir generates sustainable drug serum levels to effectively inhibit
the protease enzyme; however, saquinavir may not generate sustained le
vels necessary to inhibit the protease enzyme. Patients treated with r
itonavir, indinavir, or nelfinavir experience similar reductions in vi
ral load and increases in CD4(+) lymphocytes; smaller effects occur am
ong those treated with saquinavir. Two randomized placebo-controlled s
tudies conducted among patients with severe immune system suppression
and substantial zidovudine treatment experience demonstrated reduced H
IV disease progression and reduced mortality with PI treatment. Genoty
pic resistance to PIs occurs; the clinical relevance of resistance is
unclear. The costs of these agents including required monitoring impos
e new and substantial costs. Conclusions.-The PIs have emerged as crit
ical drugs for people with HIV infection. Optimal use involves combina
tion with reverse transcriptase inhibitors. Resistance develops to eac
h agent, and cross-resistance is likely. These agents must be used at
full doses with attention to ensuring patient compliance. The expense
of these agents may be offset by forestalling disease progression and
death and returning people to productive life. Selecting the initial P
I must be individualized, and factors to consider include proven activ
ity, possible toxicities, dosing regimens, drug interactions, and cost
s.