Recurrence or Worsening of AIDS-defining Opportunistic Infection (OI) due to Immune Reconstitution Inflammatory Syndrome (IRIS)
During Initial HAART Among a Clinic-Based Population
Poster #H-2332
Harborview Medical Center
325 9th Avenue, Box 359931
Seattle, WA, 98121
206-744-3631
[email protected]
Achenbach C and Kitahata M -- University of Washington, Seattle, WA, USA
Methods
Study type: Single center retrospective observational study conducted
among all patients in the University of Washington HIV Cohort
Patients Included:
18 years of age with CDC AIDS-defining illness
Non-infectious (wasting, dementia, lymphoma, cervical CA) and
chronic or recurrent infections difficult to classify as IRIS (recurrent
pneumonia, chronic HSV ulcer, salmonella septicemia) were excluded
OI stable or improving prior to initiation of first HAART
1 log copies/ml decline in HIV viral load
Recurrence or Worsening of OI Assessment:
During the first 12 months of HAART, records for every patient were
reviewed for recurrent or worsening signs or symptoms of prior OI
Definition of IRIS: Recurrence or worsening of OI with an atypical
inflammatory reaction based on examination, radiography, and/or
laboratory results
Statistical analyses: We performed univariate comparisons of
demographic and clinical characteristics between patients with and
without IRIS using Chi-square and Wilcoxon rank sums. Comparisons were
also made between those with recurrence or worsening due to IRIS versus
other reasons
Results
201 patients with 272 prior OIs initiated HAART between 1/96 and
12/07 (Tables 1 and 2)
Median time from OI diagnosis to HAART initiation was 2.7 months
(range 0-50)
Table 1. Demographic and clinical characteristics overall and of patients with and
without IRIS
Characteristic
Overall
(N=201)
N (%)
IRIS
(N=29)
N (%)
No-IRIS
(N=172)
N (%)
P-value
84 (42)
117 (58)
13 (45)
26 (55)
71 (41)
101 (59)
0.72
Male
Female
175 (87)
26 (13)
26 (90)
3 (10)
149 (87)
23 (13)
0.65
135 (67)
66 (33)
20 (69)
9 (31)
115 (67)
57 (33)
0.82
Among IRIS patients, 10 (34%) were hospitalized, 6 (21%) had
persistent symptoms and 2 (7%) died
Race
White
Non-white
HIV transmission risk factor
IDU
Other
57 (28)
144 (72)
7 (24)
22 (76)
50 (29)
122 (71)
0.59
CD4+ cell count nadir (cells/mm3)
<50
50
127 (63)
74 (37)
16 (55)
13 (45)
111 (65)
61 (35)
0.33
HAART started
NNRTI
PI
Boosted PI
PI + NNRTI
63 (31)
60 (30)
56 (28)
22 (11)
9 (31)
6 (21)
13 (45)
1 (3)
54 (31)
54 (31)
43 (25)
21 (12)
0.10
56 (28)
145 (72)
13 (45)
16 (55)
43 (25)
129 (75)
0.03
89 (44)
112 (56)
15 (52)
14 (48)
Table 2. OIs diagnosed prior to HAART initiation
OI
N (%)
PCP
Candida esophagitis
KS (visceral)
KS (mucocutaneous)
TB
CMV
Cryptosporidiosis
MAC
Toxoplasmosis
Cryptococcus
HSV esophagitis
PML
Histoplasmosis
Isosporiasis
74 (27)
64 (24)
7 (2)
35 (13)
28 (10)
16 (7)
14 (5)
12 (4)
9 (3)
7 (2)
2 (1)
2 (1)
1 (0.5)
1 (0.5)
Both deaths had KS (one visceral and one MC) and 4/6 (66%)
of persistent disease had MC KS
17/201 (8.5%) patients had another reason for recurrence or
worsening (Table 3): 8 with new different infection and 9 with
usual recurrence of OI (absence of atypical inflammatory reaction)
Characteristic
IRIS (N=29)
N (%)
Other (N=17)
N (%)
P-value
74 (43)
98 (57)
0.38
KS (mucocutaneous)
KS (visceral)
TB
MAC
PCP
CMV
Candida Esophagitis
Cryptococcus
Cryptosporidiosis
12 (40)
2 (7)
5 (17)
3 (10)
4 (14)
1 (4)
1 (4)
1 (4)
0 (0)
0 (0)
0 (0)
1 (6)
1 (6)
5 (29)
0 (0)
7 (41)
1 (6)
2 (12)
0.001
16 (55)
13 (45)
13 (76)
4 (24)
0.14
9 (31)
20 (69)
10 (59)
7 (41)
0.06
0 (0)
17 (100)
0.03
Event Duration (days)
<60
60
On OI treatment at HAART initiation
Yes
No
CD4+ cell count nadir (cells/mm3)
<50
50
Time from HAART initiation to IRIS (days)
OI Diagnoses
Time from OI diagnosis to HAART initiation (days)
<60
60
Hospitalization during IRIS was greatest for those with PCP
(3/4, 75%) and TB (3/5, 60%)
Table 3. Comparison of clinical characteristics between those with recurrence or
worsening due to IRIS versus other reasons
Boosted PI-based HAART
Yes
No
23/29 (79%) of IRIS events in 6 months (Figure)
IRIS was most frequent in patients with KS (14/42, 34%), MAC
(3/12, 25%), TB (5/28, 18%), and Cryptococcus (1/7, 14%) (Tables 2
and 3)
Sex
Figure. IRIS events per OI and time from HAART initiation to event
Recurrence or worsening occurred in 46/201 (23%) of patients
with 29/201 (14%) due to IRIS
Median time from initiation of HAART to IRIS was 10 weeks (range
0.7 56)
Age
<45
45
Results (cont.) OI Recurrence or Worsening on HAART
Percentage of IRIS Events
Background
There is uncertainty regarding the frequency, predictors, and outcomes of
IRIS events
Prior studies on IRIS have been limited to convenience samples of selected
OIs (usually TB, MAC, and/or cryptococcus)
We explored IRIS, in a clinic population, among all patients with any prior
OI who were started their first HAART regimen
7 (24)
22 (76)
Summary of Results
14% of all patients with a prior OI experienced IRIS on their first
HAART regimen
The majority of IRIS events occurred early in HAART (< 6
months) for all diagnoses
One-third of IRIS events led to severe disease requiring
hospitalization
KS-IRIS was notable for the following:
1/3 of cases occurred late in HAART treatment
Led to 2/2 deaths and 2/3 of persistent disease
A greater proportion of IRIS patients were treated with boosted
PI-based HAART regimens
Conclusions
In a clinic population of all patients with OIs who initiated HAART, we
observed IRIS to be a severe condition leading to hospitalization and,
occasionally, death
The type of HAART used and resulting dynamics of HIV suppression or
immunologic reactivation are likely important factors for the
development of IRIS
Further investigation of immunologic markers prior to IRIS might help
predict this serious syndrome
