- Email: [email protected]
Colistin-based treatment for extensively drug-resistant Acinetobacter baumannii pneumonia
International Journal of Antimicrobial Agents 43 (2014) 378–382
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Short Communication
Colistin-based treatment for extensively drug-resistant Acinetobacter baumannii pneumonia夽 Thana Khawcharoenporn a,∗ , Nattapol Pruetpongpun a , Pimsiri Tiamsak b , Sasinuch Rutchanawech a , Linda M. Mundy c , Anucha Apisarnthanarak a a
Division of Infectious Diseases, Faculty of Medicine, Thammasat University, Pathumthani, Thailand Thammasat University, Pathumthani, Thailand c GlaxoSmithKline, Collegeville, PA, USA b
a r t i c l e
i n f o
Article history: Received 16 October 2013 Accepted 22 January 2014 Keywords: Extensively drug-resistant Acinetobacter baumannii Pneumonia Outcomes Treatment Colistin
a b s t r a c t Data for treatment and outcomes of extensively drug-resistant Acinetobacter baumannii (XDR-AB) pneumonia are limited. A retrospective cohort study of 236 adult patients with XDR-AB pneumonia was conducted between January 2009 and December 2012. The median age of subjects was 70 years (range 17–95 years), 53% were male, 55% had ventilator-associated pneumonia and 42% had been admitted to the intensive care unit. All XDR-AB isolates were susceptible only to tigecycline and colistin; 52 (22%) of the 236 subjects did not receive an agent active against XDR-AB, with an associated 28-day survival of 0%. Colistin-based two-drug combination treatment was prescribed to 166 subjects (70%); regimens included (i) colistin and high-dose sulbactam (n = 93); (ii) colistin and tigecycline (n = 43); and (iii) colistin and high-dose prolonged infusion of a carbapenem (n = 30). The 28-day survival rate and mean length of hospital stay were not statistically different between these three regimens (65%, 53% and 60% and 39, 39 and 38 days, respectively). Predictors of mortality included Acute Physiology and Chronic Health Evaluation (APACHE) II score [adjusted odds ratio (aOR) = 1.11; P < 0.001 for each point increase], duration from infection onset to receipt of active regimen (aOR = 1.01; P = 0.002 for each hour delay), underlying malignancy (aOR = 3.46; P = 0.01) and chronic kidney disease (aOR = 2.85; P = 0.03). These findings suggest that the three colistin-based two-drug combination regimens may be treatment options for XDR-AB pneumonia. Crown Copyright © 2014 Published by Elsevier B.V. on behalf of International Society of Chemotherapy. All rights reserved.
1. Introduction Extensively drug-resistant Acinetobacter baumannii (XDR-AB) has emerged globally as a pathogen causing healthcare-associated pneumonia [1]. Previous studies have reported mortality rates of up to 65% with XDR-AB pneumonia in a variety of treated populations [2–5]. Treatment of XDR-AB pneumonia is challenging as clinical isolates are usually susceptible only to colistin [4,5]. Colistin monotherapy for pneumonia has been associated with 38–57% clinical response rates and 45–69% microbiological response rates [2–4]. A case series demonstrated a 57% clinical response rate and
86% microbiological response rate when nebulised colistin was used [6]. These poor response rates have contributed to the practice of colistin use in combination with other available antibiotics with potential synergy such as tigecycline, rifampicin, sulbactam and carbapenems [7–9]. Combined colistin and rifampicin was used successfully in 57% of patients, whilst the clinical response rates of colistin combined with the other antibiotics have not been reported for XDR-AB pneumonia [4]. This study was conducted to assess the treatment patterns and outcomes of XDR-AB pneumonia defined as either hospital-acquired pneumonia (HAP) or ventilator-associated pneumonia (VAP). 2. Methods
夽 Parts of this study were presented at the 53rd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), 10–13 September 2013, Denver, CO [poster K-175]. ∗ Corresponding author. Tel.: +66 2 926 9794; fax: +66 2 926 9793. E-mail address: [email protected] (T. Khawcharoenporn).
2.1. Study setting The study was conducted at Thammasat University Hospital (TUH) (Pathumthani, Thailand), which serves patient populations
http://dx.doi.org/10.1016/j.ijantimicag.2014.01.016 0924-8579/Crown Copyright © 2014 Published by Elsevier B.V. on behalf of International Society of Chemotherapy. All rights reserved.
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
from Pathumthani and nearby provinces in central Thailand. The study population included adult patients diagnosed with XDR-AB pneumonia from 1 January 2009 to 31 December 2012 at TUH. This study was approved by the Faculty of Medicine, Thammasat University Ethics Committee. 2.2. Study design and case definitions Eligible patients were identified by review of expectorated, induced or suctioned sputum or bronchoalveolar lavage culture results during the study period. Cases with mixed isolated microorganisms were excluded. XDR-AB was resistant to most classes of antibiotics except for two or less classes. Antibiotic susceptibility testing was performed using the Kirby–Bauer disk diffusion method with interpretation according to Clinical and Laboratory Standards Institute (CLSI) criteria. The disk diffusion breakpoints for tigecycline were ≥16 mm and ≤12 mm for susceptible and resistant, respectively, as previously reported [10]. The diagnosis of pneumonia was based on radiographic evidence of a new or progressive infiltrate, along with clinical findings of new onset of fever, purulent sputum, leukocytosis or decline in oxygenation [11]. HAP was defined as pneumonia occurring ≥48 h after hospital admission, whilst VAP was a type of HAP that developed more than 48–72 h after endotracheal intubation [11]. In this study, subjects were categorised by HAP versus VAP at the onset of infection. Subjects who died within 48 h after the onset of infection (n = 10) were excluded. Acute Physiology and Chronic Health Evaluation (APACHE) II scores were measured at the onset of infection. 2.3. Antibiotic regimens for extensively drug-resistant Acinetobacter baumannii pneumonia Antibiotic regimens prescribed after the diagnosis of XDR-AB pneumonia were at the discretion of treating physicians. An XDRAB-active regimen was a regimen that contained at least one antibiotic to which A. baumannii was susceptible. The antibiotics and dosage for normal renal function prescribed at TUH for XDR-AB pneumonia were standardised based on the protocol of the Division of Infectious Diseases, including: (i) inhaled colistin (40 mg of colistin diluted in 2 mL of sterile normal saline delivered via a mask or ventilator nebuliser every 6 h); (ii) intravenous (i.v.) colistin (300 mg loading followed by 150 mg i.v. every 12 h); (iii) tigecycline (100 mg i.v. as a loading dose followed by 50 mg i.v. every 12 h); (iv) high-dose sulbactam (6 g/day of i.v. sulbactam in the form of cefoperazone/sulbactam or ampicillin/sulbactam); and (v) high-dose prolonged i.v. infusion of carbapenems (imipenem 1 g i.v. over 3 h every 8 h, meropenem 1 g i.v. over 3 h every 8 h or doripenem 1 g i.v. over 4 h every 8 h). The dosage of colistin was adjusted according to a previous colistin dosing study [12]. The formulation of colistin was colistin methanesulfonate (1 mg equals 12 500 international units). The duration of all treatment regimens was 14 days. 2.4. Outcomes measurement Evaluation of efficacy was based on 28-day survival. Other outcomes of interest included microbiological cure at the end of therapy, defined as a follow-up sputum or bronchoalveolar lavage specimen without growth of XDR-AB at 14 days, and length of hospital stay. Adverse drug reactions were assessed as clinical signs and symptoms and via laboratory findings. Acute kidney injury (AKI) was defined according to the Kidney Disease—Improving Global Outcomes (KDIGO) Clinical Practice Guidelines. Evidence of neurotoxicity from colistin was obtained from the review of physicians’ notes.
379
2.5. Statistical analysis All analyses were performed using SPSS v.15.0 (SPSS Inc., Chicago, IL). Categorical variables were compared using the Pearson’s 2 test or Fisher’s exact test as appropriate. Continuous variables were compared using the Mann–Whitney U-test. P-values of <0.05 were considered statistically significant. Survival was compared between subjects who received different antibiotic regimens using the Kaplan–Meier estimate of the survival curve. Adjusted odd ratios (aORs) and 95% confidence intervals (CIs) were calculated in multivariate logistic regression analysis. 3. Results 3.1. Characteristics of the study population The study cohort comprised 236 subjects [median age 70 years (range 17–95 years), 125 (53%) male] and the most common comorbidity was diabetes mellitus (30%). Moreover, 100 subjects (42%) were in the intensive care unit when the XDR-AB pneumonia occurred and 130 (55%) initially had VAP (Table 1). 3.2. Definitive antibiotic regimens for extensively drug-resistant Acinetobacter baumannii pneumonia In total 184 subjects (78%) received an appropriate XDR-AB regimen and 52 (22%) received non-active (inappropriate) regimens inclusive of carbapenems (n = 31), piperacillin/tazobactam (n = 12), ceftriaxone (n = 17) and fluoroquinolones (n = 1). Compared with subjects treated with an appropriate XDR-AB regimen, those treated with a non-active XDR-AB regimen had significantly higher median APACHE II score, shorter duration of mechanical ventilation before the onset of pneumonia, receipt of steroids within 1 month prior to the onset of pneumonia and no infectious diseases consultation (Table 1). By multivariate logistic regression analysis, factors associated with receipt of non-active XDR-AB therapy were no infectious diseases consultation (aOR = 65.51, 95% CI 20.62–208.21; P < 0.001) and APACHE II score (aOR = 1.09, 95% CI 1.02–1.16; P = 0.01 for each point increase). 3.3. Treatment outcomes of colistin-based two-drug combination regimens and non-active extensively drug-resistant Acinetobacter baumannii regimens The three colistin-based two-drug combination regimens prescribed for XDR-AB pneumonia were colistin plus high-dose sulbactam (CL + SB) (n = 93), colistin plus tigecycline (CL + TC) (n = 43) and colistin plus high-dose prolonged-infusion carbapenem (CL + CB) (n = 30). For the CL + SB group, 85/93 (91%) received inhaled colistin while 8/93 (9%) received i.v. colistin. For the CL + TC group, 40/43 (93%) received inhaled colistin while 3/43 (7%) received i.v. colistin. For the CL + CB group, 28/30 (93%) received inhaled colistin while 2/30 (7%) received i.v. colistin. Characteristics of the subjects treated with these regimens were not statistically different (Table 2). The 28-day survival rates for the CL + SB, CL + TC and CL + CB groups were 65%, 53% and 60%, respectively (Table 2), and the mean survival time was comparable among the treatment groups (Fig. 1). The survival rates and the mean survival times for HAP and VAP were also comparable (Table 2). Other outcomes, including the microbiological cure rates and the median length of hospital stay after infection were not statistically different for the three groups (Table 2). The only reported adverse reaction was AKI, which occurred in all subject groups treated with i.v. colistin, at a median of 5 days into treatment with comparable rates among the three regimens. After AKI, i.v. colistin was switched to inhaled colistin with improvement in renal function for all cases.
380
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
Table 1 Characteristics of the 236 cohort patients with extensively drug-resistant Acinetobacter baumannii (XDR-AB) pneumonia categorised by treatment regimen.a Characteristic
All (N = 236)
Receipt of an active XDR-AB regimen (n = 184)
Receipt of a non-active XDR-AB regimen (n = 52)
P-valueb
Age (years) [median (range)] Male sex APACHE II score c [median (range)] Underlying diseases Diabetes mellitus Coronary artery disease Cerebrovascular disease Chronic kidney disease stage III–V Malignancy Chronic lung disease Bed-ridden status Presence of CVC Duration of CVC (days) [median (range)] Presence of urinary catheter Duration of urinary catheter (days) [median (range)] Prior surgery within 1 month Prior corticosteroid use within 1 month d Prior ICU admission within 1 month Current ICU admission Length of hospital stay e (days) [median (range)] Type of nosocomial pneumonia HAP VAP Duration of mechanical ventilation e (days) [median (range)] Infectious diseases consultation
70 (17–95) 125 (53) 21 (6–47)
75 (17–95) 101 (55) 18 (6–47)
72 (21–91) 24 (46) 25 (13–92)
0.12 0.27 <0.001*
70 (30) 36 (15) 64 (27) 39 (17) 38 (16) 36 (15) 88 (37) 72 (31) 9 (1–50) 169 (72) 10 (2–302) 48 (20) 38 (16) 22 (9) 100 (42) 12 (2–758)
60 (33) 29 (16) 49 (27) 31 (17) 26 (14) 28 (15) 69 (38) 55 (30) 9 (1–50) 130 (71) 9 (2–302) 41 (22) 21 (11) 17 (9) 84 (46) 13 (2–758)
10 (19) 7 (13) 15 (29) 8 (15) 12 (23) 8 (15) 19 (37) 17 (33) 5 (2–36) 39 (75) 7 (2–49) 7 (13) 17 (33) 5 (10) 16 (31) 11 (2–49)
0.06 0.68 0.75 0.80 0.12 0.98 0.90 0.70 0.47 0.54 0.09 0.16 < 0.001* 0.93 0.06 0.25 0.67
106 (45) 130 (55) 8 (2–758) 164 (70)
84 (46) 100 (54) 8 (2–758) 160 (87)
22 (42) 30 (58) 5 (2–31) 4 (8)
0.03* <0.001*
APACHE, Acute Physiology and Chronic Health Evaluation; CVC, central venous catheter; ICU, intensive care unit; HAP, hospital-acquired pneumonia; VAP, ventilatorassociated pneumonia. a Data are n (%) unless otherwise stated. b Comparison between patients receiving active and non-active regimens. c Measured at the onset of XDR-AB pneumonia. d Receipt of corticosteroid equivalent to prednisolone ≥10 mg/day for ≥1 month duration. e Before the onset of the infection. * Statistically significant (P < 0.05).
There was no evidence of pulmonary adverse reactions or neurotoxicity. The 28-day survival rate was 0% among the 52 subjects who received a non-active XDR-AB regimen, with a mean survival time of 4 days.
3.4. Predictors of mortality There were 125 deaths (53%) within 28 days of the onset of pneumonia. Compared with subjects who survived, subjects who died had significantly higher median APACHE II score (24 vs. 16; P < 0.001), were more likely to receive a non-active XDR-AB regimen (P < 0.001), did not have an infectious diseases consultation (P < 0.001), were more likely to have malignancy (P = 0.005) and had chronic kidney disease (P = 0.01). The 28-day mortality was similar among subjects who had HAP and VAP (53% vs. 52%). By multivariate analysis, predictors of mortality were receipt of a nonactive XDR-AB regimen (aOR = 16.95, 95% CI 6.16–50.44; P < 0.001), no infectious diseases consultation (aOR = 5.23, 95% CI 2.48–11.04; P < 0.001), APACHE II score (aOR = 1.14, 95% CI 1.08–1.20; P < 0.001 for each point increase) and malignancy (aOR = 2.44, 95% CI 1.00–6.44; P = 0.05). Among the 184 subjects who received an appropriate XDR-AB regimen, 73 (40%) died within 28 days; rates were comparable among subjects who had HAP (42%) and VAP (38%). Predictors of mortality were APACHE II score (aOR = 1.11, 95% CI 1.06–1.17; P < 0.001 for each point increase), time to receipt of active therapy (aOR = 1.01, 95% CI 1.01–1.02; P = 0.002 for each hour delay), malignancy (aOR = 3.46, 95% CI 1.35–8.88; P = 0.01) and chronic kidney disease (aOR = 2.85, 95% CI 1.12–7.26; P = 0.03).
4. Discussion Fig. 1. Kaplan–Meier survival analysis for death at 28 days after the onset of extensively drug-resistant Acinetobacter baumannii pneumonia. Comparison between treatment regimens by log-rank test: (i) CL + TC vs. CL + SB, P = 0.30; (ii) CL + TC vs. CL + CB, P = 0.53; (iii) CL + SB vs. CL + CB, P = 0.86; (iv) CL + TC vs. no active therapy, P < 0.001; (v) CL + SB vs. no active therapy, P < 0.001; and (vi) CL + CB vs. no active therapy, P < 0.001. CL, colistin; TC, tigecycline; SB, high-dose sulbactam; CB, high-dose prolonged-infusion carbapenem.
The major finding of this study was that colistin-based regimens were common for treatment of XDR-AB pneumonia at this Thai hospital (TUH) during the years 2009–2012. There are limited in vivo data for combination therapy against XDR-AB infections in the literature and this study, whilst retrospective, provides valuable, clinically relevant treatment information. Inhaled colistin was
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
381
Table 2 Characteristics and outcomes of 166 patients treated with colistin-based two-drug combination regimens for extensively drug-resistant Acinetobacter baumannii pneumonia.a Variable Age (years) [median (range)] Male sex APACHE II score c [median (range)] Underlying diseases Diabetes mellitus Coronary artery disease Cerebrovascular disease Chronic kidney disease stage III–V Malignancy Chronic lung disease Bed-ridden status Presence of CVC Duration of CVC (days) [median (range)] Presence of urinary catheter Duration of urinary catheter (days) [median (range)] Prior surgery within 1 month Prior corticosteroid use within 1 month d Prior ICU admission within 1 month Current ICU admission Length of hospital stay e (days) [median (range)] Type of nosocomial pneumonia HAP VAP Duration of mechanical ventilation e (days) [median (range)] Time to receipt of the treatment regimen f (h) [median (range)] Inhaled colistin Outcomes Survival at 28 days All pneumonia HAP VAP Mean survival time (days) by log-rank test All pneumonia HAP VAP Microbiological cure at the end of therapy Median LoS after infection (days) (range) Adverse reactions from treatment regimen g
CL + SB (n = 93)
CL + TC (n = 43)
CL + CB (n = 30)
P-value b
75 (17–92) 45 (48) 18 (6–47)
75 (45–94) 28 (65) 20 (7–41)
75 (23–88) 15 (50) 18 (8–30)
0.61 0.18 0.38
30 (32) 10 (11) 20 (22) 14 (15) 10 (11) 10 (11) 34 (37) 26 (28) 11 (1–50) 67 (72) 10 (2–302) 21 (23) 11 (12) 9 (10) 41 (44) 12 (2–758)
17 (40) 9 (21) 16 (37) 13 (30) 5 (12) 9 (21) 14 (33) 11 (26) 9 (2–29) 32 (74) 10 (3–43) 8 (19) 4 (9) 3 (7) 21 (49) 17 (3–90)
8 (27) 7 (23) 10 (33) 3 (10) 6 (20) 6 (20) 16 (53) 10 (33) 6 (2–13) 17 (57) 11 (4–50) 9 (30) 1 (3) 3 (10) 11 (37) 13 (4–60)
0.50 0.14 0.12 0.05 0.40 0.31 0.17 0.77 0.24 0.21 0.65 0.52 0.39 0.86 0.60 0.69 0.63
43 (46) 50 (54) 10 (3–758) 72 (12–240) 85 (91)
18 (42) 25 (58) 9 (3–43) 79 (1–197) 40 (93)
14 (47) 16 (53) 8 (2–60) 95 (1–192) 28 (93)
60 (65) 25/43 (58) 35/50 (70)
23 (53) 9/18 (50) 14/25 (56)
23 22 23 59/70 (84) 39 (15–172) 8 (9)
22 23 21 25/38 (66) 39 (3–137) 3 (7)
18 (60) 9/16 (56) 9/14 (64) 24 22 25 16/22 (73) 38 (11–89) 2 (7)
0.26 0.87 0.99
0.47 0.84 0.49 0.57 0.98 0.42 0.08 0.79 0.99
CL, colistin; SB, high-dose sulbactam; TC, tigecycline; CB, high-dose prolonged-infusion carbapenem; APACHE, Acute Physiology and Chronic Health Evaluation; CVC, central venous catheter; ICU, intensive care unit; HAP, hospital-acquired pneumonia; VAP, ventilator-associated pneumonia; LoS, length of hospital stay. a Data are in n (%) unless otherwise stated. b Comparison between patients receiving the three different regimens. c Measured at the onset of extensively drug-resistant Acinetobacter baumannii pneumonia. d Receipt of corticosteroid equivalent to prednisolone ≥10 mg/day for ≥1 month duration. e Duration before the onset of the infection. f Duration from the onset of infection to the initiation of treatment. g Adverse reactions were acute kidney injuries and occurred only in patients who received intravenous colistin.
the initial regimen prescribed for the majority (>90%) of subjects. The preferential selection of inhaled colistin was presumably due to less systemic toxicity compared with i.v. colistin [6]. Bronchospasm caused by inhaled colistin, whilst usually mild and reversible, was not evident in this study, nor were there reports of treatment interruption or discontinuation. Given the limited efficacy of inhaled colistin, combining inhaled colistin with at least one other active agent from another antibiotic class has become the preferred treatment for XDR-AB in this setting. The three colistin-based combination regimens were associated with 28-day survival and microbiological eradication of XDR-AB infection. Each regimen had comparable treatment effect measured by microbiological eradication, adverse drug events, length of hospital stay and survival. Notably, the higher than usual sulbactam doses and higher-dose prolonged-infusion carbapenems were prescribed to increase the time above the minimum inhibitory concentration (T > MIC) and purportedly overcome the in vitro resistance of A. baumannii. Given the comparable efficacy of these combination regimens, selection of the appropriate regimen for XDR-AB should be individualised and based on other factors including drug availability, drug intolerance, adverse reactions, potential drug–drug interactions and cost.
Of the 236 subjects, 52 (22%) did not receive appropriate XDR-AB therapy. One of the independent factors associated with receipt of non-active agents for XDR-AB was no infectious diseases consultation, suggesting that opportunities exist for physician education with regard to the local prevalence of and risk factors associated with XDR-AB infections. These risk factors included prolonged hospital stay, severe illness, device placement, colonisation with XDR-AB and exposure to group 2 carbapenems and aminopenicillins [1]. These results concur with a previous study and suggest the need for infectious diseases consultation in appropriate antibiotic selection and in the management of difficult-to-treat or multidrug-resistant bacterial infections [13]. Severity of illness measured by APACHE II score was also found to be associated with receipt of a non-active regimen for XDR-AB pneumonia. This finding may be explained by the rapid deterioration and early mortality of the subjects with high APACHE II scores before the definitive XDR-AB active regimens were given. Predictors for 28-day mortality included severity of illness and malignancy, consistent with those reported in a previous study of A. baumannii infection [14]. Receipt of non-active XDR-AB therapy and no infectious diseases consultation were predictors of mortality, whilst delay in antibiotic administration was associated
382
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
with mortality in subjects who received appropriate XDR-AB regimens. These results underlie the important role of infectious diseases specialists in the selection of early empirical and definitive active treatment and in monitoring antibiotic administration, treatment responses and adverse reactions. Chronic kidney disease was also predictive of mortality among subjects who received colistin-based regimens, which may be associated with relative immunodeficiency and vulnerability to infectious diseases of this patient population [15]. This study has some limitations. First, this was a retrospective study with recognised inherent bias for information and measurement. Second, the results were from one tertiary-care medical centre in Thailand and may not be applicable to other settings with differences in protocols and available types and administrative forms of colistin-based regimens. Third, we did not control the selection of definitive antibiotic regimens. High-dose sulbactam preference of the treating physicians may have been due to the lower cost and universal medical coverage of the drug at the study site. However, this selection bias should be minimal given the comparable patient characteristics between the three groups of colistin-based regimens. Lastly, the small sample size in each combination regimen group may have limited power to detect differences in the outcomes of interest. In conclusion, XDR-AB pneumonia was associated with significant mortality. The 28-day mortality was 100% among subjects who did not receive appropriate regimens, whilst infectious diseases consultation was an independent predictor of survival. Given that the combination of colistin with either high-dose sulbactam, tigecycline or high-dose prolonged-infusion carbapenems resulted in comparable outcomes, selection of these treatment regimens should be individualised in patients with XDR-AB pneumonia. Funding: This study was supported by the National Research University Project of the Thailand Office of Higher Education Commission (to AA and TK). Competing interests: LMM is an employee of GlaxoSmithKline (GSK) and this work was conducted pro bono and independently of GSK. All other authors declare no competing interests. Ethical approval: This study was approved by the Faculty of Medicine, Thammasat University Ethical Committee (Pathumthani, Thailand).
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
References [15] [1] Apisarnthanarak A, Hsu LY, Khawcharoenporn T, Mundy LM. Carbapenemresistant Gram-negative bacteria: how to prioritize infection prevention and
control interventions in resource-limited settings? Expert Rev Anti Infect Ther 2013;11:147–57. Garnacho-Montero J, Ortiz-Leyba C, Jiménez-Jiménez FJ, Barrero-Almodóvar AE, García-Garmendia JL, Bernabeu-WittelI M. Treatment of multidrugresistant Acinetobacter baumannii ventilator-associated pneumonia (VAP) with intravenous colistin: a comparison with imipenem-susceptible VAP. Clin Infect Dis 2003;36:1111–8. Kalin G, Alp E, Coskun R, Demiraslan H, Gündogan K, Doganay M. Use of high-dose IV and aerosolized colistin for the treatment of multidrug-resistant Acinetobacter baumannii ventilator-associated pneumonia: do we really need this treatment? J Infect Chemother 2012;18:872–7. Durante-Mangoni E, Signoriello G, Andini R, Mattei A, De Cristoforo M, Murino P, et al. Colistin and rifampicin compared with colistin alone for the treatment of serious infections due to extensively drug-resistant Acinetobacter baumannii: a multicenter, randomized clinical trial. Clin Infect Dis 2013;57: 349–58. Lee YT, Kuo SC, Yang SP, Lin YT, Chiang DH, Tseng FC, et al. Bacteremic nosocomial pneumonia caused by Acinetobacter baumannii and Acinetobacter nosocomialis: a single or two distinct clinical entities? Clin Microbiol Infect 2013;19:640–5. Kwa AL, Loh C, Low JG, Kurup A, Tam VH. Nebulized colistin in the treatment of pneumonia due to multidrug-resistant Acinetobacter baumannii and Pseudomonas aeruginosa. Clin Infect Dis 2005;41:754–7. Mutlu Yilmaz E, Sunbul M, Aksoy A, Yilmaz H, Guney AK, Guvenc T. Efficacy of tigecycline/colistin combination in a pneumonia model caused by extensively drug-resistant Acinetobacter baumannii. Int J Antimicrob Agents 2012;40:332–6. ˜ Pachón-Ibánez ME, Docobo-Pérez F, López-Rojas R, Domínguez-Herrera J, Jiménez-Mejias ME, García-Curiel A, et al. Efficacy of rifampin and its combinations with imipenem, sulbactam, and colistin in experimental models of infection caused by imipenem-resistant Acinetobacter baumannii. Antimicrob Agents Chemother 2010;54:1165–72. Pongpech P, Amornnopparattanakul S, Panapakdee S, Fungwithaya S, Nannha P, Dhiraputra C, et al. Antibacterial activity of carbapenem-based combinations against multidrug-resistant Acinetobacter baumannii. J Med Assoc Thai 2010;93:161–71. Jones RN, Ferraro MJ, Reller LB, Schreckenberger PC, Swenson JM, Sader HS. Multicenter studies of tigecycline disk diffusion susceptibility results for Acinetobacter spp. J Clin Microbiol 2007;45:227–30. American Thoracic Society; Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilatorassociated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005;171:388–416. Garonzik SM, Li J, Thamlikitkul V, Paterson DL, Shoham S, Jacob J, et al. Population pharmacokinetics of colistin methanesulfonate and formed colistin in critically ill patients from a multicenter study provide dosing suggestions for various categories of patients. Antimicrob Agents Chemother 2011;55:3284–94. Esposito S, Leone S. Antimicrobial treatment for intensive care unit (ICU) infections including the role of the infectious disease specialist. Int J Antimicrob Agents 2007;29:494–500. Anunnatsiri S, Tonsawan P. Risk factors and clinical outcomes of multidrug-resistant Acinetobacter baumannii bacteremia at a university hospital in Thailand. Southeast Asian J Trop Med Public Health 2011;42: 693–703. Sarnak MJ, Jaber BL. Mortality caused by sepsis in patients with end-stage renal disease compared with the general population. Kidney Int 2000;58: 1758–64.
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Short Communication
Colistin-based treatment for extensively drug-resistant Acinetobacter baumannii pneumonia夽 Thana Khawcharoenporn a,∗ , Nattapol Pruetpongpun a , Pimsiri Tiamsak b , Sasinuch Rutchanawech a , Linda M. Mundy c , Anucha Apisarnthanarak a a
Division of Infectious Diseases, Faculty of Medicine, Thammasat University, Pathumthani, Thailand Thammasat University, Pathumthani, Thailand c GlaxoSmithKline, Collegeville, PA, USA b
a r t i c l e
i n f o
Article history: Received 16 October 2013 Accepted 22 January 2014 Keywords: Extensively drug-resistant Acinetobacter baumannii Pneumonia Outcomes Treatment Colistin
a b s t r a c t Data for treatment and outcomes of extensively drug-resistant Acinetobacter baumannii (XDR-AB) pneumonia are limited. A retrospective cohort study of 236 adult patients with XDR-AB pneumonia was conducted between January 2009 and December 2012. The median age of subjects was 70 years (range 17–95 years), 53% were male, 55% had ventilator-associated pneumonia and 42% had been admitted to the intensive care unit. All XDR-AB isolates were susceptible only to tigecycline and colistin; 52 (22%) of the 236 subjects did not receive an agent active against XDR-AB, with an associated 28-day survival of 0%. Colistin-based two-drug combination treatment was prescribed to 166 subjects (70%); regimens included (i) colistin and high-dose sulbactam (n = 93); (ii) colistin and tigecycline (n = 43); and (iii) colistin and high-dose prolonged infusion of a carbapenem (n = 30). The 28-day survival rate and mean length of hospital stay were not statistically different between these three regimens (65%, 53% and 60% and 39, 39 and 38 days, respectively). Predictors of mortality included Acute Physiology and Chronic Health Evaluation (APACHE) II score [adjusted odds ratio (aOR) = 1.11; P < 0.001 for each point increase], duration from infection onset to receipt of active regimen (aOR = 1.01; P = 0.002 for each hour delay), underlying malignancy (aOR = 3.46; P = 0.01) and chronic kidney disease (aOR = 2.85; P = 0.03). These findings suggest that the three colistin-based two-drug combination regimens may be treatment options for XDR-AB pneumonia. Crown Copyright © 2014 Published by Elsevier B.V. on behalf of International Society of Chemotherapy. All rights reserved.
1. Introduction Extensively drug-resistant Acinetobacter baumannii (XDR-AB) has emerged globally as a pathogen causing healthcare-associated pneumonia [1]. Previous studies have reported mortality rates of up to 65% with XDR-AB pneumonia in a variety of treated populations [2–5]. Treatment of XDR-AB pneumonia is challenging as clinical isolates are usually susceptible only to colistin [4,5]. Colistin monotherapy for pneumonia has been associated with 38–57% clinical response rates and 45–69% microbiological response rates [2–4]. A case series demonstrated a 57% clinical response rate and
86% microbiological response rate when nebulised colistin was used [6]. These poor response rates have contributed to the practice of colistin use in combination with other available antibiotics with potential synergy such as tigecycline, rifampicin, sulbactam and carbapenems [7–9]. Combined colistin and rifampicin was used successfully in 57% of patients, whilst the clinical response rates of colistin combined with the other antibiotics have not been reported for XDR-AB pneumonia [4]. This study was conducted to assess the treatment patterns and outcomes of XDR-AB pneumonia defined as either hospital-acquired pneumonia (HAP) or ventilator-associated pneumonia (VAP). 2. Methods
夽 Parts of this study were presented at the 53rd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), 10–13 September 2013, Denver, CO [poster K-175]. ∗ Corresponding author. Tel.: +66 2 926 9794; fax: +66 2 926 9793. E-mail address: [email protected] (T. Khawcharoenporn).
2.1. Study setting The study was conducted at Thammasat University Hospital (TUH) (Pathumthani, Thailand), which serves patient populations
http://dx.doi.org/10.1016/j.ijantimicag.2014.01.016 0924-8579/Crown Copyright © 2014 Published by Elsevier B.V. on behalf of International Society of Chemotherapy. All rights reserved.
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
from Pathumthani and nearby provinces in central Thailand. The study population included adult patients diagnosed with XDR-AB pneumonia from 1 January 2009 to 31 December 2012 at TUH. This study was approved by the Faculty of Medicine, Thammasat University Ethics Committee. 2.2. Study design and case definitions Eligible patients were identified by review of expectorated, induced or suctioned sputum or bronchoalveolar lavage culture results during the study period. Cases with mixed isolated microorganisms were excluded. XDR-AB was resistant to most classes of antibiotics except for two or less classes. Antibiotic susceptibility testing was performed using the Kirby–Bauer disk diffusion method with interpretation according to Clinical and Laboratory Standards Institute (CLSI) criteria. The disk diffusion breakpoints for tigecycline were ≥16 mm and ≤12 mm for susceptible and resistant, respectively, as previously reported [10]. The diagnosis of pneumonia was based on radiographic evidence of a new or progressive infiltrate, along with clinical findings of new onset of fever, purulent sputum, leukocytosis or decline in oxygenation [11]. HAP was defined as pneumonia occurring ≥48 h after hospital admission, whilst VAP was a type of HAP that developed more than 48–72 h after endotracheal intubation [11]. In this study, subjects were categorised by HAP versus VAP at the onset of infection. Subjects who died within 48 h after the onset of infection (n = 10) were excluded. Acute Physiology and Chronic Health Evaluation (APACHE) II scores were measured at the onset of infection. 2.3. Antibiotic regimens for extensively drug-resistant Acinetobacter baumannii pneumonia Antibiotic regimens prescribed after the diagnosis of XDR-AB pneumonia were at the discretion of treating physicians. An XDRAB-active regimen was a regimen that contained at least one antibiotic to which A. baumannii was susceptible. The antibiotics and dosage for normal renal function prescribed at TUH for XDR-AB pneumonia were standardised based on the protocol of the Division of Infectious Diseases, including: (i) inhaled colistin (40 mg of colistin diluted in 2 mL of sterile normal saline delivered via a mask or ventilator nebuliser every 6 h); (ii) intravenous (i.v.) colistin (300 mg loading followed by 150 mg i.v. every 12 h); (iii) tigecycline (100 mg i.v. as a loading dose followed by 50 mg i.v. every 12 h); (iv) high-dose sulbactam (6 g/day of i.v. sulbactam in the form of cefoperazone/sulbactam or ampicillin/sulbactam); and (v) high-dose prolonged i.v. infusion of carbapenems (imipenem 1 g i.v. over 3 h every 8 h, meropenem 1 g i.v. over 3 h every 8 h or doripenem 1 g i.v. over 4 h every 8 h). The dosage of colistin was adjusted according to a previous colistin dosing study [12]. The formulation of colistin was colistin methanesulfonate (1 mg equals 12 500 international units). The duration of all treatment regimens was 14 days. 2.4. Outcomes measurement Evaluation of efficacy was based on 28-day survival. Other outcomes of interest included microbiological cure at the end of therapy, defined as a follow-up sputum or bronchoalveolar lavage specimen without growth of XDR-AB at 14 days, and length of hospital stay. Adverse drug reactions were assessed as clinical signs and symptoms and via laboratory findings. Acute kidney injury (AKI) was defined according to the Kidney Disease—Improving Global Outcomes (KDIGO) Clinical Practice Guidelines. Evidence of neurotoxicity from colistin was obtained from the review of physicians’ notes.
379
2.5. Statistical analysis All analyses were performed using SPSS v.15.0 (SPSS Inc., Chicago, IL). Categorical variables were compared using the Pearson’s 2 test or Fisher’s exact test as appropriate. Continuous variables were compared using the Mann–Whitney U-test. P-values of <0.05 were considered statistically significant. Survival was compared between subjects who received different antibiotic regimens using the Kaplan–Meier estimate of the survival curve. Adjusted odd ratios (aORs) and 95% confidence intervals (CIs) were calculated in multivariate logistic regression analysis. 3. Results 3.1. Characteristics of the study population The study cohort comprised 236 subjects [median age 70 years (range 17–95 years), 125 (53%) male] and the most common comorbidity was diabetes mellitus (30%). Moreover, 100 subjects (42%) were in the intensive care unit when the XDR-AB pneumonia occurred and 130 (55%) initially had VAP (Table 1). 3.2. Definitive antibiotic regimens for extensively drug-resistant Acinetobacter baumannii pneumonia In total 184 subjects (78%) received an appropriate XDR-AB regimen and 52 (22%) received non-active (inappropriate) regimens inclusive of carbapenems (n = 31), piperacillin/tazobactam (n = 12), ceftriaxone (n = 17) and fluoroquinolones (n = 1). Compared with subjects treated with an appropriate XDR-AB regimen, those treated with a non-active XDR-AB regimen had significantly higher median APACHE II score, shorter duration of mechanical ventilation before the onset of pneumonia, receipt of steroids within 1 month prior to the onset of pneumonia and no infectious diseases consultation (Table 1). By multivariate logistic regression analysis, factors associated with receipt of non-active XDR-AB therapy were no infectious diseases consultation (aOR = 65.51, 95% CI 20.62–208.21; P < 0.001) and APACHE II score (aOR = 1.09, 95% CI 1.02–1.16; P = 0.01 for each point increase). 3.3. Treatment outcomes of colistin-based two-drug combination regimens and non-active extensively drug-resistant Acinetobacter baumannii regimens The three colistin-based two-drug combination regimens prescribed for XDR-AB pneumonia were colistin plus high-dose sulbactam (CL + SB) (n = 93), colistin plus tigecycline (CL + TC) (n = 43) and colistin plus high-dose prolonged-infusion carbapenem (CL + CB) (n = 30). For the CL + SB group, 85/93 (91%) received inhaled colistin while 8/93 (9%) received i.v. colistin. For the CL + TC group, 40/43 (93%) received inhaled colistin while 3/43 (7%) received i.v. colistin. For the CL + CB group, 28/30 (93%) received inhaled colistin while 2/30 (7%) received i.v. colistin. Characteristics of the subjects treated with these regimens were not statistically different (Table 2). The 28-day survival rates for the CL + SB, CL + TC and CL + CB groups were 65%, 53% and 60%, respectively (Table 2), and the mean survival time was comparable among the treatment groups (Fig. 1). The survival rates and the mean survival times for HAP and VAP were also comparable (Table 2). Other outcomes, including the microbiological cure rates and the median length of hospital stay after infection were not statistically different for the three groups (Table 2). The only reported adverse reaction was AKI, which occurred in all subject groups treated with i.v. colistin, at a median of 5 days into treatment with comparable rates among the three regimens. After AKI, i.v. colistin was switched to inhaled colistin with improvement in renal function for all cases.
380
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
Table 1 Characteristics of the 236 cohort patients with extensively drug-resistant Acinetobacter baumannii (XDR-AB) pneumonia categorised by treatment regimen.a Characteristic
All (N = 236)
Receipt of an active XDR-AB regimen (n = 184)
Receipt of a non-active XDR-AB regimen (n = 52)
P-valueb
Age (years) [median (range)] Male sex APACHE II score c [median (range)] Underlying diseases Diabetes mellitus Coronary artery disease Cerebrovascular disease Chronic kidney disease stage III–V Malignancy Chronic lung disease Bed-ridden status Presence of CVC Duration of CVC (days) [median (range)] Presence of urinary catheter Duration of urinary catheter (days) [median (range)] Prior surgery within 1 month Prior corticosteroid use within 1 month d Prior ICU admission within 1 month Current ICU admission Length of hospital stay e (days) [median (range)] Type of nosocomial pneumonia HAP VAP Duration of mechanical ventilation e (days) [median (range)] Infectious diseases consultation
70 (17–95) 125 (53) 21 (6–47)
75 (17–95) 101 (55) 18 (6–47)
72 (21–91) 24 (46) 25 (13–92)
0.12 0.27 <0.001*
70 (30) 36 (15) 64 (27) 39 (17) 38 (16) 36 (15) 88 (37) 72 (31) 9 (1–50) 169 (72) 10 (2–302) 48 (20) 38 (16) 22 (9) 100 (42) 12 (2–758)
60 (33) 29 (16) 49 (27) 31 (17) 26 (14) 28 (15) 69 (38) 55 (30) 9 (1–50) 130 (71) 9 (2–302) 41 (22) 21 (11) 17 (9) 84 (46) 13 (2–758)
10 (19) 7 (13) 15 (29) 8 (15) 12 (23) 8 (15) 19 (37) 17 (33) 5 (2–36) 39 (75) 7 (2–49) 7 (13) 17 (33) 5 (10) 16 (31) 11 (2–49)
0.06 0.68 0.75 0.80 0.12 0.98 0.90 0.70 0.47 0.54 0.09 0.16 < 0.001* 0.93 0.06 0.25 0.67
106 (45) 130 (55) 8 (2–758) 164 (70)
84 (46) 100 (54) 8 (2–758) 160 (87)
22 (42) 30 (58) 5 (2–31) 4 (8)
0.03* <0.001*
APACHE, Acute Physiology and Chronic Health Evaluation; CVC, central venous catheter; ICU, intensive care unit; HAP, hospital-acquired pneumonia; VAP, ventilatorassociated pneumonia. a Data are n (%) unless otherwise stated. b Comparison between patients receiving active and non-active regimens. c Measured at the onset of XDR-AB pneumonia. d Receipt of corticosteroid equivalent to prednisolone ≥10 mg/day for ≥1 month duration. e Before the onset of the infection. * Statistically significant (P < 0.05).
There was no evidence of pulmonary adverse reactions or neurotoxicity. The 28-day survival rate was 0% among the 52 subjects who received a non-active XDR-AB regimen, with a mean survival time of 4 days.
3.4. Predictors of mortality There were 125 deaths (53%) within 28 days of the onset of pneumonia. Compared with subjects who survived, subjects who died had significantly higher median APACHE II score (24 vs. 16; P < 0.001), were more likely to receive a non-active XDR-AB regimen (P < 0.001), did not have an infectious diseases consultation (P < 0.001), were more likely to have malignancy (P = 0.005) and had chronic kidney disease (P = 0.01). The 28-day mortality was similar among subjects who had HAP and VAP (53% vs. 52%). By multivariate analysis, predictors of mortality were receipt of a nonactive XDR-AB regimen (aOR = 16.95, 95% CI 6.16–50.44; P < 0.001), no infectious diseases consultation (aOR = 5.23, 95% CI 2.48–11.04; P < 0.001), APACHE II score (aOR = 1.14, 95% CI 1.08–1.20; P < 0.001 for each point increase) and malignancy (aOR = 2.44, 95% CI 1.00–6.44; P = 0.05). Among the 184 subjects who received an appropriate XDR-AB regimen, 73 (40%) died within 28 days; rates were comparable among subjects who had HAP (42%) and VAP (38%). Predictors of mortality were APACHE II score (aOR = 1.11, 95% CI 1.06–1.17; P < 0.001 for each point increase), time to receipt of active therapy (aOR = 1.01, 95% CI 1.01–1.02; P = 0.002 for each hour delay), malignancy (aOR = 3.46, 95% CI 1.35–8.88; P = 0.01) and chronic kidney disease (aOR = 2.85, 95% CI 1.12–7.26; P = 0.03).
4. Discussion Fig. 1. Kaplan–Meier survival analysis for death at 28 days after the onset of extensively drug-resistant Acinetobacter baumannii pneumonia. Comparison between treatment regimens by log-rank test: (i) CL + TC vs. CL + SB, P = 0.30; (ii) CL + TC vs. CL + CB, P = 0.53; (iii) CL + SB vs. CL + CB, P = 0.86; (iv) CL + TC vs. no active therapy, P < 0.001; (v) CL + SB vs. no active therapy, P < 0.001; and (vi) CL + CB vs. no active therapy, P < 0.001. CL, colistin; TC, tigecycline; SB, high-dose sulbactam; CB, high-dose prolonged-infusion carbapenem.
The major finding of this study was that colistin-based regimens were common for treatment of XDR-AB pneumonia at this Thai hospital (TUH) during the years 2009–2012. There are limited in vivo data for combination therapy against XDR-AB infections in the literature and this study, whilst retrospective, provides valuable, clinically relevant treatment information. Inhaled colistin was
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
381
Table 2 Characteristics and outcomes of 166 patients treated with colistin-based two-drug combination regimens for extensively drug-resistant Acinetobacter baumannii pneumonia.a Variable Age (years) [median (range)] Male sex APACHE II score c [median (range)] Underlying diseases Diabetes mellitus Coronary artery disease Cerebrovascular disease Chronic kidney disease stage III–V Malignancy Chronic lung disease Bed-ridden status Presence of CVC Duration of CVC (days) [median (range)] Presence of urinary catheter Duration of urinary catheter (days) [median (range)] Prior surgery within 1 month Prior corticosteroid use within 1 month d Prior ICU admission within 1 month Current ICU admission Length of hospital stay e (days) [median (range)] Type of nosocomial pneumonia HAP VAP Duration of mechanical ventilation e (days) [median (range)] Time to receipt of the treatment regimen f (h) [median (range)] Inhaled colistin Outcomes Survival at 28 days All pneumonia HAP VAP Mean survival time (days) by log-rank test All pneumonia HAP VAP Microbiological cure at the end of therapy Median LoS after infection (days) (range) Adverse reactions from treatment regimen g
CL + SB (n = 93)
CL + TC (n = 43)
CL + CB (n = 30)
P-value b
75 (17–92) 45 (48) 18 (6–47)
75 (45–94) 28 (65) 20 (7–41)
75 (23–88) 15 (50) 18 (8–30)
0.61 0.18 0.38
30 (32) 10 (11) 20 (22) 14 (15) 10 (11) 10 (11) 34 (37) 26 (28) 11 (1–50) 67 (72) 10 (2–302) 21 (23) 11 (12) 9 (10) 41 (44) 12 (2–758)
17 (40) 9 (21) 16 (37) 13 (30) 5 (12) 9 (21) 14 (33) 11 (26) 9 (2–29) 32 (74) 10 (3–43) 8 (19) 4 (9) 3 (7) 21 (49) 17 (3–90)
8 (27) 7 (23) 10 (33) 3 (10) 6 (20) 6 (20) 16 (53) 10 (33) 6 (2–13) 17 (57) 11 (4–50) 9 (30) 1 (3) 3 (10) 11 (37) 13 (4–60)
0.50 0.14 0.12 0.05 0.40 0.31 0.17 0.77 0.24 0.21 0.65 0.52 0.39 0.86 0.60 0.69 0.63
43 (46) 50 (54) 10 (3–758) 72 (12–240) 85 (91)
18 (42) 25 (58) 9 (3–43) 79 (1–197) 40 (93)
14 (47) 16 (53) 8 (2–60) 95 (1–192) 28 (93)
60 (65) 25/43 (58) 35/50 (70)
23 (53) 9/18 (50) 14/25 (56)
23 22 23 59/70 (84) 39 (15–172) 8 (9)
22 23 21 25/38 (66) 39 (3–137) 3 (7)
18 (60) 9/16 (56) 9/14 (64) 24 22 25 16/22 (73) 38 (11–89) 2 (7)
0.26 0.87 0.99
0.47 0.84 0.49 0.57 0.98 0.42 0.08 0.79 0.99
CL, colistin; SB, high-dose sulbactam; TC, tigecycline; CB, high-dose prolonged-infusion carbapenem; APACHE, Acute Physiology and Chronic Health Evaluation; CVC, central venous catheter; ICU, intensive care unit; HAP, hospital-acquired pneumonia; VAP, ventilator-associated pneumonia; LoS, length of hospital stay. a Data are in n (%) unless otherwise stated. b Comparison between patients receiving the three different regimens. c Measured at the onset of extensively drug-resistant Acinetobacter baumannii pneumonia. d Receipt of corticosteroid equivalent to prednisolone ≥10 mg/day for ≥1 month duration. e Duration before the onset of the infection. f Duration from the onset of infection to the initiation of treatment. g Adverse reactions were acute kidney injuries and occurred only in patients who received intravenous colistin.
the initial regimen prescribed for the majority (>90%) of subjects. The preferential selection of inhaled colistin was presumably due to less systemic toxicity compared with i.v. colistin [6]. Bronchospasm caused by inhaled colistin, whilst usually mild and reversible, was not evident in this study, nor were there reports of treatment interruption or discontinuation. Given the limited efficacy of inhaled colistin, combining inhaled colistin with at least one other active agent from another antibiotic class has become the preferred treatment for XDR-AB in this setting. The three colistin-based combination regimens were associated with 28-day survival and microbiological eradication of XDR-AB infection. Each regimen had comparable treatment effect measured by microbiological eradication, adverse drug events, length of hospital stay and survival. Notably, the higher than usual sulbactam doses and higher-dose prolonged-infusion carbapenems were prescribed to increase the time above the minimum inhibitory concentration (T > MIC) and purportedly overcome the in vitro resistance of A. baumannii. Given the comparable efficacy of these combination regimens, selection of the appropriate regimen for XDR-AB should be individualised and based on other factors including drug availability, drug intolerance, adverse reactions, potential drug–drug interactions and cost.
Of the 236 subjects, 52 (22%) did not receive appropriate XDR-AB therapy. One of the independent factors associated with receipt of non-active agents for XDR-AB was no infectious diseases consultation, suggesting that opportunities exist for physician education with regard to the local prevalence of and risk factors associated with XDR-AB infections. These risk factors included prolonged hospital stay, severe illness, device placement, colonisation with XDR-AB and exposure to group 2 carbapenems and aminopenicillins [1]. These results concur with a previous study and suggest the need for infectious diseases consultation in appropriate antibiotic selection and in the management of difficult-to-treat or multidrug-resistant bacterial infections [13]. Severity of illness measured by APACHE II score was also found to be associated with receipt of a non-active regimen for XDR-AB pneumonia. This finding may be explained by the rapid deterioration and early mortality of the subjects with high APACHE II scores before the definitive XDR-AB active regimens were given. Predictors for 28-day mortality included severity of illness and malignancy, consistent with those reported in a previous study of A. baumannii infection [14]. Receipt of non-active XDR-AB therapy and no infectious diseases consultation were predictors of mortality, whilst delay in antibiotic administration was associated
382
T. Khawcharoenporn et al. / International Journal of Antimicrobial Agents 43 (2014) 378–382
with mortality in subjects who received appropriate XDR-AB regimens. These results underlie the important role of infectious diseases specialists in the selection of early empirical and definitive active treatment and in monitoring antibiotic administration, treatment responses and adverse reactions. Chronic kidney disease was also predictive of mortality among subjects who received colistin-based regimens, which may be associated with relative immunodeficiency and vulnerability to infectious diseases of this patient population [15]. This study has some limitations. First, this was a retrospective study with recognised inherent bias for information and measurement. Second, the results were from one tertiary-care medical centre in Thailand and may not be applicable to other settings with differences in protocols and available types and administrative forms of colistin-based regimens. Third, we did not control the selection of definitive antibiotic regimens. High-dose sulbactam preference of the treating physicians may have been due to the lower cost and universal medical coverage of the drug at the study site. However, this selection bias should be minimal given the comparable patient characteristics between the three groups of colistin-based regimens. Lastly, the small sample size in each combination regimen group may have limited power to detect differences in the outcomes of interest. In conclusion, XDR-AB pneumonia was associated with significant mortality. The 28-day mortality was 100% among subjects who did not receive appropriate regimens, whilst infectious diseases consultation was an independent predictor of survival. Given that the combination of colistin with either high-dose sulbactam, tigecycline or high-dose prolonged-infusion carbapenems resulted in comparable outcomes, selection of these treatment regimens should be individualised in patients with XDR-AB pneumonia. Funding: This study was supported by the National Research University Project of the Thailand Office of Higher Education Commission (to AA and TK). Competing interests: LMM is an employee of GlaxoSmithKline (GSK) and this work was conducted pro bono and independently of GSK. All other authors declare no competing interests. Ethical approval: This study was approved by the Faculty of Medicine, Thammasat University Ethical Committee (Pathumthani, Thailand).
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
References [15] [1] Apisarnthanarak A, Hsu LY, Khawcharoenporn T, Mundy LM. Carbapenemresistant Gram-negative bacteria: how to prioritize infection prevention and
control interventions in resource-limited settings? Expert Rev Anti Infect Ther 2013;11:147–57. Garnacho-Montero J, Ortiz-Leyba C, Jiménez-Jiménez FJ, Barrero-Almodóvar AE, García-Garmendia JL, Bernabeu-WittelI M. Treatment of multidrugresistant Acinetobacter baumannii ventilator-associated pneumonia (VAP) with intravenous colistin: a comparison with imipenem-susceptible VAP. Clin Infect Dis 2003;36:1111–8. Kalin G, Alp E, Coskun R, Demiraslan H, Gündogan K, Doganay M. Use of high-dose IV and aerosolized colistin for the treatment of multidrug-resistant Acinetobacter baumannii ventilator-associated pneumonia: do we really need this treatment? J Infect Chemother 2012;18:872–7. Durante-Mangoni E, Signoriello G, Andini R, Mattei A, De Cristoforo M, Murino P, et al. Colistin and rifampicin compared with colistin alone for the treatment of serious infections due to extensively drug-resistant Acinetobacter baumannii: a multicenter, randomized clinical trial. Clin Infect Dis 2013;57: 349–58. Lee YT, Kuo SC, Yang SP, Lin YT, Chiang DH, Tseng FC, et al. Bacteremic nosocomial pneumonia caused by Acinetobacter baumannii and Acinetobacter nosocomialis: a single or two distinct clinical entities? Clin Microbiol Infect 2013;19:640–5. Kwa AL, Loh C, Low JG, Kurup A, Tam VH. Nebulized colistin in the treatment of pneumonia due to multidrug-resistant Acinetobacter baumannii and Pseudomonas aeruginosa. Clin Infect Dis 2005;41:754–7. Mutlu Yilmaz E, Sunbul M, Aksoy A, Yilmaz H, Guney AK, Guvenc T. Efficacy of tigecycline/colistin combination in a pneumonia model caused by extensively drug-resistant Acinetobacter baumannii. Int J Antimicrob Agents 2012;40:332–6. ˜ Pachón-Ibánez ME, Docobo-Pérez F, López-Rojas R, Domínguez-Herrera J, Jiménez-Mejias ME, García-Curiel A, et al. Efficacy of rifampin and its combinations with imipenem, sulbactam, and colistin in experimental models of infection caused by imipenem-resistant Acinetobacter baumannii. Antimicrob Agents Chemother 2010;54:1165–72. Pongpech P, Amornnopparattanakul S, Panapakdee S, Fungwithaya S, Nannha P, Dhiraputra C, et al. Antibacterial activity of carbapenem-based combinations against multidrug-resistant Acinetobacter baumannii. J Med Assoc Thai 2010;93:161–71. Jones RN, Ferraro MJ, Reller LB, Schreckenberger PC, Swenson JM, Sader HS. Multicenter studies of tigecycline disk diffusion susceptibility results for Acinetobacter spp. J Clin Microbiol 2007;45:227–30. American Thoracic Society; Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilatorassociated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005;171:388–416. Garonzik SM, Li J, Thamlikitkul V, Paterson DL, Shoham S, Jacob J, et al. Population pharmacokinetics of colistin methanesulfonate and formed colistin in critically ill patients from a multicenter study provide dosing suggestions for various categories of patients. Antimicrob Agents Chemother 2011;55:3284–94. Esposito S, Leone S. Antimicrobial treatment for intensive care unit (ICU) infections including the role of the infectious disease specialist. Int J Antimicrob Agents 2007;29:494–500. Anunnatsiri S, Tonsawan P. Risk factors and clinical outcomes of multidrug-resistant Acinetobacter baumannii bacteremia at a university hospital in Thailand. Southeast Asian J Trop Med Public Health 2011;42: 693–703. Sarnak MJ, Jaber BL. Mortality caused by sepsis in patients with end-stage renal disease compared with the general population. Kidney Int 2000;58: 1758–64.