Transcript Style F 24 by 48

O-219
Reasons for Hospitalizations that Occur after HAART Initiation
CROI
Johns Hopkins University School of Medicine, Baltimore, MD, USA
Table 1. Baseline Characteristics
Age at HAART
Background: Hospitalization rates for AIDS defining illnesses (ADI) decrease after HAART initiation.
hospitalization rates for other diagnoses change after HAART.
It is unclear how
Methods: This is a prospective observational study of 2031 HIV-infected patients followed by the Johns Hopkins AIDS Service who
first initiated HAART 1996 – 2005. Hospitalization rates and causes were examined before and after first HAART initiation.
Modified Clinical Classification Software 2008 was used to assign the primary cause of each hospitalization to 1 of 18 diagnostic
categories. Relative hospitalization rates were estimated using negative binomial regression.
Results: The cohort was 34% female, 76% Black, 45% IDU, 27% MSM and had a median age of 39 years (IQR 34, 44). About half
(54%) initiated HAART in 1996 to 1998. The median CD4 count and HIV-1 RNA level at HAART initiation were 152 cells/mm3
(37, 298) and 4.8 Log10 copies/mL (4.1, 5.3) respectively. Thirty-eight percent of subjects had > 1 hospitalization in the overall
period from 180 days prior to HAART through 365 days afterward. Admissions due to any cause declined by 46 – 90 days after
HAART (RR vs. 180 days prior, 0.76 [0.62, 0.93]).
Non-ADI infections, ADI, psychiatric, GI and liver, endocrine / metabolic, renal, and cardiovascular were the seven most
common diagnostic categories. Compared to the 180 days prior to HAART (Figure 2), statistically significant decreases in
hospitalization rates were seen at 91 – 180 days after HAART for ADI (RR, 0.60 [0.42, 0.85]) and psychiatric illness (0.47 [0.26,
0.83]), and at 181 – 365 days for ADI (0.46 [0.33, 0.63]), psychiatric illness (0.60 [0.37, 0.98]), and non-ADI infections (0.72 [0.55,
0.94]). Virologic data to determine whether a subject had a significant response to HAART (> 1Log10 decrease in HIV-1 RNA within
6 months) were available for 1327 subjects. Among this group, virologic responders showed significant decreases in ADI and nonADI infections after 90 days while non-responders did not.
Conclusions: Hospitalization is common, and infections (both ADI and non-ADI) are the reasons for the majority of hospitalizations
within the period surrounding HAART initiation. Providers should continue to be aware of the high risk of infectious illness through
at least one year following HAART initiation, although decreases in risk of both infectious and psychiatric admissions may occur
within 3 to 6 months.
Background
Figure 1. All-cause
Hospitalization Rate Prior to
and After HAART Initiation
* *
*
Hospitalization rates were calculated per 100
person years (PY) for the periods: 180 days prior
to HAART initiation and days 1–45, 46–90, 91180, and 181–365 after HAART. Multivariate
negative binomial regression with generalized
estimating equations was used to estimate relative
hospitalization rates and standard errors.
For the 1327 persons with sufficient data, virologic
response was defined as a decrease in HIV-1 RNA
of >1 Log10 at 6 months after the HAART
initiation. Among virologic responders, immune
reconstitution inflammatory syndrome (IRIS) was
identified by chart review.
<50 cells / mm3
576 (28)
30-39
878 (43)
50-199
623 (31)
40-49
691 (34)
200-349
460 (23)
>50
224 (11)
>350
372 (18)
39 (33.7, 44.1)
Gender
Median (IQR) in cells /
mm3
A. 965 Virologic Responders
*
*
HIV RNA at HAART
688 (34)
<4 Log10 copies/mL
415 (20)
Men
1343 (66)
4-5
765 (38)
>5
851 (42)
Racial / Ethnic Category
Median (IQR) in Log10 copies/mL
African American (AA)
1550 (76)
White
438 (22)
**
*
152 (37, 298)
Women
*
*
4.8 (4.1, 5.3)
HAART Type
Hispanic (non-AA)
20 (1)
NNRTI (plus >2 NRTI’s)
566 (28)
Asian
6 (<1)
PI (plus >2 NRTI’s)
1238 (61)
Other
17 (1)
PI and NNRTI (plus >1 NRTI)
181 (9)
> 3 NRTI’s (w/o PI or NNRTI)
46 (2)
* Indicates P<0.05 for the relative hospitalization rate compared to the rate 180 days prior to HAART initiation.
*
*
*
*
B. 362 Non-Responders
*
Interpretation: Hospitalizations for Non-AIDS defining infections, AIDS defining illnesses and psychiatric illnesses decreased statistically
significantly within the year after HAART initiation, while hospitalizations for other illnesses did not.
*
IDU
435 (21)
IDU-Heterosexual
391 (19)
1996-1998
1093 (54)
89 (4)
1999-2002
603 (30)
Heterosexual
539 (27)
2003-2005
335 (17)
MSM
457 (23)
Unknown / Other
120 (6)
Calendar Era at HAART
Table 3.. Multivariate Analysis of Factors Associated with
Hospitalization Rates
All Cause
Prior 180 days
Table 2. Top 3 Reasons for Hospitalization by Category
No. (% of all
admissions)
Non-AIDS Defining
Infections
459 (26)
AIDS Defining
Illnesses
406 (23)
Psychiatric
200 (11)
Gastrointestinal and
Liver
92 (5)
Endocrine,
Nutritional, Metabolic,
and Immunity
82 (5)
Renal and
Genitourinary
76 (4)
ADI
Pneumonia – organism NOS (16%), bacterial endocarditis (6%), cellulitis
of leg (6%)
1.00 (ref)
1.00 (ref)
1.00 (ref)
1.00 (ref)
Days 1-45
0.99 (0.82, 1.19)
1.07 (0.77, 1.49)
1.03 (0.75, 1.42)
0.64 (0.35, 1.17)
Days 46-90
0.75 (0.61, 0.92)
0.82 (0.56, 1.20)
0.63 (0.43, 0.94)
0.68 (0.37, 1.25)
Days 91-180
0.75 (0.62, 0.90)
0.75 (0.53, 1.04)
0.55 (0.39, 0.78)
0.50 (0.28, 0.88)
Days 181-365
0.73 (0.62, 0.86)
0.73 (0.56, 0.96)
0.40 (0.29, 0.55)
0.64 (0.40, 1.04)
Age at HAART > 39 years
1.18 (1.00, 1.40)
1.33 (1.06, 1.68)
---
0.60 (0.39, 0.93)
Women
1.41 (1.20, 1.68)
1.68 (1.34, 2.12)
---
1.58 (1.05, 2.37)
African American
1.25 (1.00, 1.58)
1.85 (1.31, 2.61)
---
1.23 (0.69, 2.22)
IDU
1.37 (1.16, 1.62)
1.52 (1.20, 1.94)
---
2.91 (1.90, 4.48)
Pneumocystosis (25%), cryptococcosis (17%), candidal esophagitis (11%)
Recurrent major depression (26%), depressive disorder NEC (14%), druginduced depression (12%)
CD4 count at HAART
---
Acute pancreatitis (13%), chronic pancreatitis (8%), cirrhosis of liver NOS
(4%), gastrointestinal hemorrhage NOS (4%)
<50 cells/mm3
2.80 (2.22, 3.53)
2.62 (1.90, 3.60)
10.42 (6.45, 16.83)
50-199
1.43 (1.13, 1.81)
1.62 (1.17, 2.25)
2.73 (1.61, 4.62)
Hypovolemia (37%), cachexia (7%), hypercalcemia (6%)
>200
1.00 (ref)
1.00 (ref)
1.00 (ref)
Acute renal failure NOS (22%), hypertensive renal failure (18%), lower
nephron nephrosis (9%)
Venous thrombosis (15%), stroke (9%), congestive heart failure (9%)
64 (4)
Chemotherapy visit (42%), reticulosarcoma NOS (6%), metastasis to
bone (6%)
Injury and Poisoning
60 (3)
Complication of dialysis (13%) or vascular (7%) device, poisoning –
antidepressant (5%)
Pulmonary
50 (3)
Asthma (52%), spontaneous pneumothorax NEC (6%), chronic bronchitis
(4%)
45 (3)
Anemia NOS (22%), aplastic anemias NEC (20%), agranulocytosis (11%)
Diagnostic categories which each had <2% of all admissions included symptomatic classification, neurologic,
obstetric and gynecologic, orthopedic, dermatologic, congenital conditions (0%), and perinatal conditions (0%).
<4 Log10 copies/mL
* Indicates P<0.05 for the relative hospitalization rate compared to the rate 180 days prior to HAART initiation.
Hatched region indicates portion of ADI’s considered immune reconstitution inflammatory syndromes.
HIV-1 RNA at HAART
74 (4)
Psych
Most Common ICD9-based Diagnoses1 (percent of the category)
Non-AIDS Defining
Neoplasms
Hematologic
Non-ADI Inf
Time with respect to HAART
All values are number of patients (% of total) unless otherwise specified. IDU, injection drug user;
MSM, men who have sex with men; NNRTI, non-nucleoside reverse transcriptase inhibitor; NRTI,
nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
Category
*
*
HIV Risk Factors
Circulatory
* Indicates P<0.05 for the relative hospitalization
rate compared to the rate 180 days prior to
HAART initiation.
Figure 3. Hospitalization Rates by Virologic Response
CD4 Count at HAART
238 (12)
IDU-MSM
HIV infection increases morbidity from opportunistic infections and other AIDS defining illnesses (ADI).
While HAART may decrease risk for many illnesses over the long-term, disease rates following initiation
are unknown. Our main objective was to measure hospitalization rates due to various causes over time in
the year after HAART initiation.
Figure 2. Hospitalization Rates by Category Prior to and After HAART
18-29 years
Median (IQR) in years
Fax: 410-537-7266
Email: [email protected]
Hospitalizations / 100 PY
Abstract
Hospitalizations prior to and after HAART
initiation were examined for all HAART naïve
patients in the Johns Hopkins HIV Clinical Cohort
who initiated HAART 1996 – 2005. Professional
abstractors assigned primary reasons for
hospitalization by ICD-9 code. Modified Clinical
Classifications Software (AHRQ, 2008) was used
to classify reasons for hospitalization into 18
categories.
1830 E. Monument Street / Suite 452
Baltimore, MD 21287
Phone: 410-502-8829
Stephen A Berry, Yukari C Manabe, Richard D Moore, Kelly A Gebo
Montreal, QC
February 8 – 11, 2009
Methods
Stephen A. Berry, M.D.
1.00 (ref)
1.00 (ref)
4-5
0.94 (0.74, 1.19)
1.25 (0.87, 1.81)
0.87 (0.56, 1.37)
>5
1.27 (0.99, 1.63)
1.81 (1.25, 2.60)
1.28 (0.84, 1.94)
1996-1998
1.00 (ref)
1.00 ref
1.00 (ref)
1999-2002
1.13 (0.94, 1.36)
1.22 (0.93, 1.58)
0.99 (0.71, 1.37)
2003-2005
1.20 (0.96, 1.51)
1.35 (0.99, 1.83)
0.76 (0.49, 1.18)
Conclusions
 All-cause hospitalization rate remains high (comparable to the pre-HAART rate) for the first 45 days
after HAART initiation, then falls significantly between 45 and 90 days after initiation
 In our cohort, Non-AIDS defining infections (Non-ADI Inf) and AIDS defining illnesses (ADI) are the
top causes of hospitalizations prior to HAART initiation and throughout the following year
 Hospitalization rates for both Non-ADI Inf and ADI remain high for 45 days then decrease; this pattern
appears particularly true for virologic responders
 Hospitalization rates for psychiatric illness also fall significantly after HAART initiation, but it may be
---
1.00 (ref)
Interpretation: The decrease in infectious and psychiatric hospitalizations after HAART initiation occurs
primarily among virologic responders. For virologic responders during days 1-45 after HAART initiation,
immune reconstitution inflammatory syndromes account for >50% of hospitalizations due to AIDS defining
illnesses and 13% of all hospitalizations.
that treatment of psychiatric illness leads to HAART initiation
 Age, sex, race and IDU are associated with hospitalization risk for Non-ADI Inf and psychiatric illness
whereas hospitalization risk for AIDS defining illnesses is most associated with CD4 count
Calendar era at HAART
---
--- Indicates not included in multivariate model because bivariate P>0.20.
HAART type (NNRTI plus >2 NRTI’s, PI plus >2 NRTI’s, PI & NNRTI plus >1 NRTI), and >3 NRTI’s) was not associated with
admissions in any category in bivariate analyses.
Implications
Clinically meaningful recovery of immune function becomes evident between 45 and 90 days after HAART
initiation. Providers should maintain close clinical vigilance for opportunistic and non-opportunistic
infections through at least 90 days even among virologic responders.