Triple combination antiviral drug for pandemic H1N1 in critically ill patients on mechanical ...

35 Pages·2011·0.5 MB·English

AAC Accepts, published online ahead of print on 3 October 2011
Antimicrob. Agents Chemother. doi:10.1128/AAC.05529-11
Copyright © 2011, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.
1 Triple combination antiviral drug for pandemic H1N1 in critically ill patients on
2 mechanical ventilation
3 Running title: Triple Therapy in Critically Ill H1N1 Infection
4
5 †Won-Young Kim1*, †Gee Young Suh2, Jin Won Huh1, Sung-Han Kim3, Min-ju Kim4,
D
o
6 Yun Seong Kim5, Hye-Ryoun Kim6, Yon Ju Ryu7, Min Soo Han8, Young Gwan Ko9, Gyu w
n
7 Rak Chon10, Kwan Ho Lee11, Sang-Ho Choi3, and Sang-Bum Hong1*; for the Korean loa
d
e
8 Society of Critical Care Medicine (KSCCM) H1N1 Collaborative‡ d

f
r
9 1Department of Pulmonary and Critical Care Medicine, 3Department of Infectious Diseases, om

h
10 and 4Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, t
t
p
:
11 University of Ulsan College of Medicine, Seoul; 2Division of Pulmonary and Critical Care //a
a
c
12 Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul; .a
s
13 5Pusan National University School of Medicine, Pusan; 6Korea Cancer Center Hospital, m
.
o
r
14 Seoul; 7Ewha Womans University School of Medicine, Seoul; 8Eulji University School of g/

o
15 Medicine, Daejon; 9School of Medicine, Kyung Hee University, Seoul; 10Konkuk University n
A
p
16 School of Medicine, Cheungju; 11College of Medicine, Yeungnam University, Daegu, ril
1
4
17 Republic of Korea ,

2
0
18 † Dr. Won-Young Kim and Dr. Gee Young Suh contributed to this paper equally. 1
9

b
19 y

g
20 ‡ Korean Society of Critical Care Medicine (KSCCM) H1N1 Collaborative ue
s
t
21 Gee Young Suh, Kyeongman Jeon (Sungkyunkwan University School of Medicine); Shin Ok
22 Koh, Moo Suk Park (Yonsei University College of Medicine); Won-Il Choi (Keimyung
23 University School of Medicine); Younsuck Koh, Chae-Man Lim, Sang-Bum Hong, Eun
24 Young Choi, Sung-Han Kim, Jaemin Lim, Jong-Joon Ahn, Jin Won Huh, Sang-Ho Choi,
1
25 Won-Young Kim, Sung-Cheol Yun, Min-ju Kim (University of Ulsan College of Medicine);
26 Hyun-Kyung Lee (College of Medicine, Inje University); Seok Chan Kim (The Catholic
27 University of Korea College of Medicine); Sang Hyun Kwak (Chonnam National University
28 Medical School); Young Joo Lee (Ajou University School of Medicine); Heung-Bum Lee
29 (Chonbuk National University Medical School); Jae Yeol Kim (Chung Ang University
D
o
30 College of Medicine); Jae Hwa Cho (Inha University College of Medicine); Hye Sook Choi w
n
lo
31 (Dongguk University Gyeongju Hospital); Myung Goo Lee, Yong Bum Park (Hallym a
d
e
32 University School of Medicine); Ho Cheol Kim (Gyongsang National University School of d

f
r
o
33 Medicine); Yeon Sook Kim (Chungnam National University Hospital); Chang Young Lim m

h
34 (CHA University College of medicine); Ki Uk Kim, Yun Seong Kim (Pusan National t
t
p
:
/
35 University School of Medicine) Hye-Ryoun Kim (Korea Cancer Center Hospital); Yon Ju /a
a
c
36 Ryu (Ewha Womans University School of Medicine); Min Soo Han (Eulji University School .a
s
m
37 of Medicine); Young Gwan Ko (School of Medicine, Kyung Hee University); Sun-Jong Kim,
.
o
r
38 Gyu Rak Chon (Konkuk University School of Medicine); Kwan Ho Lee (College of g/

o
n
39 Medicine, Yeungnam University).
A
p
40 ril
1
4
41 Corresponding author: Sang-Bum Hong, MD, Department of Pulmonary and Critical Care ,

2
0
42 Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 1
9

b
43 43-Gil, Songpa-gu, Seoul 138-736, Republic of Korea. E-mail: [email protected] y

g
44 Tel: [phone] Fax: [phone] ue
s
t
45
46
47
48
2
49 ABSTRACT
50 Recent in vitro study showed that the three compounds of antiviral drugs with different
51 mechanisms of action (amantadine, ribavirin, and oseltamivir) could result in synergistic
52 antiviral activity. However, no clinical studies have evaluated the efficacy and safety of
53 combination antiviral therapy in patients with severe influenza illness. 245 adult patients who
D
o
54 were critically ill with confirmed pandemic influenza A/H1N1 2009 (pH1N1) infection and w
n
lo
55 were admitted to one of the intensive care units (ICUs) of 28 hospitals in Korea were a
d
e
56 reviewed. Patients who required ventilator support and received either triple combination d

f
r
o
57 antiviral drug (TCAD) therapy or oseltamivir monotherapy were analyzed. A total of 127 m

h
58 patients were included in our analysis. Among them, 24 patients received TCAD therapy and t
t
p
:
/
59 103 patients received oseltamivir monotherapy. The 14-day mortality was 17% in the TCAD /a
a
c
60 group and 35% in the oseltamivir group (P = 0.08) and the 90-day mortality was 46% in the .a
s
m
61 TCAD group and 59% in the oseltamivir group (P = 0.23). None of the toxicities attributable
.
o
r
62 to antiviral drugs occurred in either group of our study, including hemolytic anemia and g/

o
n
63 hepatic toxicities related with the use of ribavirin. Logistic regression analysis indicated that
A
p
64 the odds ratio (OR) for the association of TCAD with 90-day mortality was 0.58 (95% CI ril
1
4
65 0.24-1.42, P = 0.24). Although this study was retrospective and did not provide virologic ,

2
0
66 outcomes, our results suggest that the treatment outcome of the triple combination of 1
9

b
67 amantadine, ribavirin, and oseltamivir was comparable to that of oseltamivir monotherapy. y

g
68 ue
s
t
69 Keywords: Pandemic A/H1N1 2009, triple combination antiviral drug, critically ill patients,
70 mechanical ventilation
71
72
3
73 INTRODUCTION
74 Pandemic influenza A/H1N1 2009 (pH1N1) was a considerable public health concern
75 worldwide. Most infected people experienced mild and uncomplicated illnesses, but some
76 patients developed rapidly progressive pneumonia that lead to respiratory failure, shock, and
77 death (6, 22, 30). The Centers for Disease Control and Prevention (CDC) recommended that
D
o
78 all patients with severe pH1N1 illness be treated with oseltamivir or zanamivir as soon as w
n
lo
79 possible (9). Oseltamivir and zanamivir are neuraminidase inhibitors (NAIs), which interfere a
d
e
80 with the release of progeny influenza virus from infected cells, thereby preventing new d

f
r
o
81 rounds of infection (27). Oseltamivir is readily available in oral formulation, and zanamivir in m

h
82 inhalation or intravenous form. Rimantadine and amantadine are closely related adamantanes t
t
p
:
/
83 (also called M2 inhibitors), and they target the M2 protein of influenza A, which forms a /a
a
c
84 proton channel in the viral membrane that is essential for efficient viral replication (35). .a
s
m
85 Ribavirin is approved for the treatment of respiratory syncytial virus and in combination with
.
o
r
86 interferon or peginterferon for hepatitis C virus. In vitro inhibitory activity against both viral g/

o
n
87 RNA and DNA polymerase may implicate for a reduced potential of ribavirin for the drug
A
p
88 resistance of influenza viruses (26). Ribavirin is available in oral, aerosolized, and ril
1
4
89 intravenous formulations, although intravenous form is not currently approved in the United ,

2
0
90 States. 1
9

b
91 Previous studies on antiviral therapies in influenza virus were conducted primarily in y

g
92 healthy outpatients with uncomplicated illnesses (12, 29). Limited data are available on ue
s
t
93 antiviral use in patients with severe influenza infection, and recommendations are mostly
94 based on expert opinions or observational studies (24, 42). Treatment can be complicated if
95 NAI-resistant viral strains are suspected (4, 40). Recent studies reported the emergence of
96 oseltamivir-resistant viruses after short-term drug therapy (16, 25). The prevalence of drug-
4
97 resistant strains could undermine clinical benefit of antiviral drugs when utilized as
98 monotherapy, especially for critically ill patients, when the development of antiviral
99 resistance is rapid.
100 The present limitations of monotherapy to treat severe influenza illness have renewed
101 interest in antiviral therapies that combine multiple drugs with different mechanisms of action
D
o
102 (15, 28, 32). Nguyen et al. evaluated the three compounds of antiviral drugs (amantadine, w
n
lo
103 ribavirin, and oseltamivir), so called a triple combination antiviral drug (TCAD) regimen, a
d
e
104 using an in vitro infection model (28). They reported that TCAD had a strong effect on drug- d

f
r
o
105 resistant viruses, including pH1N1 strains. Such combination therapy might be clinically m

h
106 useful in treatment of influenza viruses that are resistant to one or more antivirals. The CDC t
t
p
:
/
107 reported that 1,143 of 1,148 (99.6%) seasonal H1N1 viruses isolated in the period 2008-2009 /a
a
c
108 and 10 of 1,497 (0.6%) pH1N1 isolates tested in the United States were resistant to .a
s
m
109 oseltamivir (4). In Korea, from May 2009 through January 2010, a total of 740,835 patients
.
o
r
110 were reported with pH1N1 virus infection and 11 of 67 (16%) patients who were suspected of g/

o
n
111 having drug-resistant pH1N1 strains were identified as oseltamivir-resistant (31). To date,
A
p
112 rapid diagnostic tests to determine the susceptibility profile of influenza viruses are not ril
1
4
113 available. The availability of a broad-spectrum antiviral therapy that would be effective ,

2
0
114 regardless of the susceptibility against each antiviral drug would be of high clinical utility. 1
9

b
115 However, no clinical studies have demonstrated the efficacy and safety of TCAD for patients y

g
116 with severe influenza illness. ue
s
t
117 In this study, we analyzed the efficacy and safety of TCAD for patients with pH1N1
118 infection whose illnesses were severe enough for admission to an ICU with ventilator support.
119
120
5
121 MATERIALS AND METHODS
122 Patient selection
123 We initially reviewed the data of a cohort from the Korean Society of Critical Care
124 Medicine (KSCCM) H1N1 Collaborative. The cohort is composed of critically ill adult
125 patients who were at least 15 years old, had confirmed pH1N1 infection, and were admitted
D
o
126 to one of the ICUs of the 28 participating tertiary or referral hospitals of Korea from w
n
lo
127 September 2009 to February 2010. Pandemic influenza A/H1N1 2009 virus was confirmed a
d
e
128 by a positive result from a probe-based reverse-transcriptase polymerase-chain-reaction test d

f
r
o
129 from a nasopharyngeal swab or broncho-alveolar lavage (37). From this cohort, we included m

h
130 in analysis patients who: (i) required ventilator support (invasive or noninvasive); (ii) t
t
p
:
/
131 received either TCAD or oseltamivir monotherapy as initial antiviral treatment. /a
a
c
132 .a
s
m
133 Antiviral treatment protocol
.
o
r
134 During 2009 pandemic, the KSCCM established the management protocol for new H1N1 g/

o
n
135 influenza pneumonia with respiratory failure (21). The KSCCM also invited the study
A
p
136 investigators to participate in this study, and encouraged them to follow the protocol. With ril
1
4
137 respect to the use of antivirals, the protocol states that triple combination antiviral regimen ,

2
0
138 (oseltamivir 150 mg twice daily + amantadine 100 mg twice daily + ribavirin 300 mg three 1
9

b
139 times daily) may be considered in case of severe influenza illness. The drug dosage for y

g
140 amantadine was based on data from product labels (www.drugs.com/pro/symmetrel.html), ue
s
t
141 and a reduction in dosage was recommended for patients with age ≥65 years or creatinine
142 clearance (CCr) <50 mL/min. The dosage for oral ribavirin was based on previously
143 published studies (5, 23), and the drug was used with caution in patients with CCr <50
144 mL/min. As for oseltamivir, we suggested doubling the dose (150 mg twice daily) when used
6
145 in triple combination regimen to treat severe influenza illness. For patients with CCr <30
146 mL/min, a reduction in dosage was recommended. The antiviral drugs were administered via
147 a nasogastric (NG) or gastric tube when the patients were on ventilator support.
148
149 Study design
D
o
150 We performed a retrospective cohort analysis to compare the clinical outcomes of patients w
n
lo
151 given TCAD therapy or oseltamivir therapy. The primary outcome of this study was mortality. a
d
e
152 To assess the safety of antiviral agents, we identified any possible adverse events related to d

f
r
o
153 antiviral therapy and reviewed serum liver enzymes and creatinine at the time of ICU m

h
154 admission (baseline) and at 3, 7, and 14 days after initiation of drugs. We also analyzed t
t
p
:
/
155 factors associated with mortality, including TCAD treatment. This study was approved by the /a
a
c
156 local Institutional Review Board (IRB) of each hospital. Informed consent was not necessary .a
s
m
157 because this was not an interventional study (IRB of Asan Medical Center: 20110294).
.
o
r
158 g/

o
n
159 Definitions
A
p
160 A nosocomial infection of pH1N1 was defined as one in which symptoms or signs ril
1
4
161 suggestive of influenza illness newly developed three days or more after hospital admission ,

2
0
162 and pH1N1 virus was confirmed as a causative agent. Cardiovascular disease included 1
9

b
163 hypertension, ischemic heart disease, arrhythmia, and congestive heart failure. An y

g
164 immunosuppressive condition was diagnosed if there was an underlying disease that affected ue
s
t
165 the immune system (chronic liver disease, chronic renal disease, human immunodeficiency
166 virus infection, or malignancy) or if immunosuppressive therapy was being administered at
167 the time of infection. Acute respiratory distress syndrome (ARDS) was diagnosed by
168 consensus definition (3). Severity of illness was assessed by the Acute Physiology and
7
169 Chronic Health Evaluation (APACHE) II score (20) and the Sequential Organ Failure
170 Assessment (SOFA) score (39) on the day of ICU admission.
171
172 Statistical analysis
173 Data are presented as the number ± standard deviation (SD) or (%) of patients unless
D
o
174 indicated otherwise. Student’s t-test or the Kruskall-Wallis test was used to compare w
n
lo
175 continuous data and the chi-square or Fisher’s exact test was used to compare categorical data a
d
e
176 as appropriate. In order to identify factors associated with 90-day mortality, all prespecified d

f
r
o
177 covariables, which are listed in Table 1 and Table 2 of the article, were included in the m

h
178 multiple logistic regression analysis by using step-wise backward selection procedures with P t
t
p
:
/
179 < 0.10. To prevent multicollinearity, variables with high correlation between each other were /a
a
c
180 strictly controlled. Model calibration was assessed with Hosmer-Lemeshow test (χ2 = 4.07, df .a
s
m
181 = 8, P = 0.85). All tests of significance were two-tailed and P value less than 0.05 were
.
o
r
182 considered significant. All analyses were performed with SPSS version 18.0K for Windows g/

o
n
183 (SPSS Inc, Chicago, IL, USA).
A
p
184 ril
1
4
185 ,

2
0
186 RESULTS 1
9

b
187 There were 245 patients in the initial cohort, and 127 of these patients were ultimately y

g
188 included in our analysis. Considerable drop-out occurred largely due to patients being treated ue
s
t
189 with NAIs other than oseltamivir (n = 32). A total of 24 of the included patients received
190 TCAD and 103 received oseltamivir monotherapy (Figure 1).
191 Table 1 shows the baseline characteristics of the TCAD group and the oseltamivir group.
192 Patients given TCAD were older (mean age, 63.5 ± 18.9 years vs. 55.7 ± 17.7 years, P =
8
193 0.046), but there was no difference in sex or body mass index. There were no differences
194 between the two groups in co-morbidities, immunosuppressive conditions, severity of illness,
195 and baseline values measured at ICU admission. The proportion of patients with shock and
196 ARDS were also similar between the two groups. Adjuvant corticosteroids were administered
197 to about half of the patients in each group. There were no differences between the groups in
D
o
198 mean time from onset of symptoms to antiviral drug administration. In the TCAD group, w
n
lo
199 higher doses of oseltamivir were administered (oseltamivir ≥300 mg/d, 63% vs. 33%, P = a
d
e
200 0.01). The mean treatment duration was longer (11.1 ± 7.0 days vs. 6.2 ± 3.6 days, P < 0.001) d

f
r
o
201 in the TCAD group, although in patients who survived (n = 55), there was no significant m

h
202 difference (9.6 ± 5.4 days vs. 7.3 ± 3.1 days, P = 0.15). The mean doses of amantadine and t
t
p
:
/
203 ribavirin in the TCAD group were less than the recommended doses of the KSCCM protocol. /a
a
c
204 Table 2 shows the clinical outcomes of the two groups. There were three cases (13%) of .a
s
m
205 community-acquired pneumonia in the TCAD group, and ten cases (10%) in the oseltamivir
.
o
r
206 group. During the course of illness, hospital-acquired pneumonia occurred and the proportion g/

o
n
207 of patients was similar between the two groups. Within 90 days from onset of symptoms, 11
A
p
208 patients (46%) died in the TCAD group and 61 (59%) in the oseltamivir group. The deaths ril
1
4
209 occurred within 14-day period in 36% (4/11) of the TCAD group and 59% (36/61) of the ,

2
0
210 oseltamivir group. There was a trend toward lower 14-day, 30-day, and 90-day mortality in 1
9

b
211 the TCAD group, but this was not statistically significant. In the TCAD group, all 11 patients y

g
212 died due to aggravation of septic shock or pneumonia, and 50 patients (82%) in the ue
s
t
213 oseltamivir group. In the oseltamivir group, three patients died due to uncontrolled liver
214 failure, two due to intra-abdominal infection, two due to exacerbation of underlying disease,
215 and four due to undetermined cause. There were no differences between the two groups in
216 mean ventilator-free days.
9
217 The results of additional analysis of the study group stratified by duration of ICU stay are
218 shown in Table 3. Most of the baseline characteristics were similar between the two groups,
219 but patients with ICU stay more than 7 days were more immunocompromised (34% vs. 52%,
220 P = 0.04) and had longer mean duration of symptom before antiviral administration (3.5 ± 3.1
221 days vs. 6.3 ± 5.1 days, P = 0.001). Secondary bacterial (hospital-acquired) pneumonia
D
o
222 occurred more in the group with longer ICU stay (31% vs. 69%, P < 0.001), however more w
n
lo
223 deaths occurred in patients with ICU stay less than or equal to 7 days (66% vs. 48%, P = a
d
e
224 0.04). d

f
r
o
225 Table 4 shows the safety measures related to antiviral agents. There were neither m

h
226 hematologic (eg. hemolytic anemia) nor neurologic (eg. seizure) events in any of the 127 t
t
p
:
/
227 patients. In the TCAD group, the mean difference (post-treatment value – baseline value) for /a
a
c
228 aspartate transaminase (AST) was -12.8 ± 38.7 IU/L, for alanine transaminase (ALT) was - .a
s
m
229 2.9 ± 21.3 IU/L, and for total bilirubin (TB) was -0.2 ± 3.0 mg/dL. In the oseltamivir group,
.
o
r
230 seven cases had severe liver enzyme elevation (Supplementary Table 1) and the mean g/

o
n
231 differences of AST and ALT were very high. There was no significant difference in the mean
A
p
232 difference of serum creatinine between the two groups. Three patients, all in oseltamivir ril
1
4
233 group, died due to uncontrolled liver failure, although the relationship between toxicity and ,

2
0
234 drug use was unclear. 1
9

b
235 Table 5 shows the results of our univariate and multivariate analysis of risk factors y

g
236 associated with 90-day mortality. APACHE II score and mean arterial pressure were excluded ue
s
t
237 in further analysis due to multicollinearity with SOFA score and shock, respectively.
238 Multivariate analysis that adjusted for variables associated with 90-day mortality rate
239 indicated that increasing age, nosocomial infection of the virus, higher SOFA score, and
240 prone position were significantly associated with increased mortality. On the other hand,
10