Original Article
1079
Efficacy and Safety of Intravenous Colistin in Preterm Infants with Nosocomial Sepsis Caused by Acinetobacter baumannii Serdar Alan, MD1 Emel Okulu, MD1
Duran Yildiz, MD1 Omer Erdeve, MD1 Ufuk Cakir, MD1 Dilek Kahvecioglu, MD1 Can Ates, MSc2 Begum Atasay, MD1 Saadet Arsan, MD1
1 Division of Neonatology, Department of Pediatrics, Ankara University
School of Medicine, Ankara, Turkey 2 Depatrment of Biostatistics, Ankara University School of Medicine, Ankara, Turkey
Address for correspondence Serdar Alan, MD, Division of Neonatology, Department of Pediatrics, Ankara University School of Medicine, Vedat Dalokay Cad. No: 90 A/16 Gaziosmanpasa, 06100 Ankara, Turkey (e-mail: [email protected]).
Abstract
Keywords
► Acinetobacter baumannii ► colistin ► neonatal intensive care unit ► nosocomial infection ► preterm
Objectives To describe the efficacy of intravenous colistin on clinical and microbiological outcomes in preterm infants with nosocomial sepsis in neonatal intensive care unit (NICU) and define adverse events observed with this treatment. Methods The records of preterm infants who received colistin with or without positive cultures in the NICU were retrospectively reviewed. Patients were evaluated for response to therapy and side effects. Results A total of 21 preterm infants with medians of 28 weeks (23–36) gestational age and 870 g (620–2,650) birth weight were included. The median duration and dose of colistin therapy were 9 days (3–26) and 3 mg/kg/d (2–5). Recovery rate in patients including all with/without positive culture was 81% (17/21). Microbiological clearance by colistin was 69% (9/13). The major side effect observed was acute kidney injury (19%). At least 24% of infants required electrolyte supplementation during the colistin therapy. Magnesium levels were significantly lower at the end of the colistin therapy (p < 0.001). Acute kidney injury and electrolyte disturbances including hypomagnesemia were reversible in all surviving patients. Conclusion We suggest that renal function tests and serum electrolytes should be monitored closely and replaced in case of any need during the colistin therapy in preterm infants.
The emergence of multidrug-resistant (MDR) nosocomial gram-negative bacteria such as Acinetobacter baumannii (A. baumannii) has become a major problem with increasing morbidity and mortality in critically ill patients including preterm infants. Meanwhile, no new antimicrobial drug to treat MDR gram-negative bacterial infection has been presented recently.1–3 Colistin which has been in use over 60 years was one of the first antibiotics with significant activity against gram-negative bacteria. However, it was largely replaced by aminoglycosides in 1970s because of
the concern about its nephrotoxicity and neurotoxicity.3 The worldwide emergence of MDR nosocomial gram-negative pathogens over the last two decades has resulted in increased use of colistin as a “salvage” therapy.4 Nephrotoxicity and neurotoxicity were reported in approximately 22 and 4% of patients, respectively, in a recent multicenter large series study in pediatric patients who were exposed to colistin for more than 72 hours.5 In contrast, two recently published trials in newborns suggested that nephrotoxicity and neurotoxicity associated with colistin were less
received January 2, 2014 accepted after revision January 21, 2014 published online February 28, 2014
Copyright © 2014 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1371361. ISSN 0735-1631.
Downloaded by: Collections and Technical Services Department. Copyrighted material.
Am J Perinatol 2014;31:1079–1086.
Efficacy and Safety of Intravenous Colistin in Preterm Infants frequent.6,7 However, there is still only scant information on intravenous colistin use especially in preterm patients.6–9 Furthermore, optimal dose and interval for safety and efficacy remain unclear. Therefore, we described the effectiveness and safety of toxicity of intravenous colistin therapy in preterm infants with and without MDR A. baumannii infection during the last 20 months.
Methods Study Design and Settings This was a retrospective study of critically ill preterm infants with proven or suspected nosocomial infection caused by MDR A. baumannii and treated with intravenous colistin. The study was conducted at a level III neonatal intensive care unit (NICU) with a capacity of 20 incubators which took care of high-risk infants of 2,500 annual inborn deliveries and highrisk referrals from all parts of Turkey. Approximately 400 patients per year were admitted to the unit. The trial was approved by the Human Research Ethics Committee of Ankara University School of Medicine.
Patient Identification and Inclusion/Exclusion Criteria All preterm patients hospitalized between January 2012 and August 2013 in the NICU and treated with intravenous colistin were evaluated. Preterm infants born before 37 weeks of gestation and treated with intravenous colistin for at least 72 hours were included in the study. Patients with major congenital anomalies were excluded.
Data Collection The medical records of all cases were reviewed from hospital files. Recorded variables included: demographic characteristics (prenatal features, gestational age [GA], birth weight [BW], sex, mode of delivery), hospital stay, postnatal age, and weight when colistin started, duration of colistin therapy, cumulative dose, concomitant drugs, and clinical condition. Blood, urine, cerebrospinal fluid, and other sterile body fluid (peritoneal fluid, etc.) cultures, and antimicrobial sensitivity were all recorded. Microbiological clearance and clinical outcomes after colistin therapy were evaluated. In addition, renal (creatinine and blood urea nitrogen level), and liver (alanine aminotransferase, aspartate aminotransferase (AST), gama glutamil transferase, and bilirubin levels) function tests and serum electrolytes (calcium, magnesium [Mg], potassium, sodium) were recorded for possible side effects before, during, and after the colistin therapy. Use of additional nephrotoxic agents that included amphotericin, aminoglycosides, diuretics, vancomycin, nonsteroidal anti-inflammatory medications, and intravenous contrast were also recorded. Short-term neurological outcomes including apnea and seizure, and long-term neurological outcomes including visual evoked potential (VEP), brainstem auditory evoked response (BAER), and neuromotor deficits were collected from patient files.
Alan et al.
and Prevention.10 MDR was defined as resistance to at least three antimicrobial agents (of different groups) with intrinsic activity against gram-negative pathogen. A favorable outcome was defined as culture negativity in culture-positive infections or clinical response with normalized acute phase reactants in clinical septic patients. The AKIN (Acute Kidney Injury Network) criteria11 were employed as a validated tool to evaluate acute kidney injury (AKI). Other clinical conditions related to AKI were defined as symptomatic patent ductus arteriosus, hypoxic ischemic birth, septic shock, hypovolemic shock, due to major bleeding, major surgery, and concomitant antimicrobial agents, including vancomycin, aminoglycosides, amphotericin B were all checked from patient files.
Microbiologic Methods BACTEC 9240 (Becton-Dickinson, NJ) was used to identify the microorganism. The VITEK 2 susceptibility card AST-N090 (bioMérieux, Marcy l’Etoile, France) containing a colistin susceptibility test was used according to the manufacturer instructions. Interpretive breakpoints (minimum inhibitory concentration [MIC] 0.5 μg/mL, susceptible, and MIC 16 μg/mL, resistant) were used for the VITEK 2.
Pharmacodynamics A course of intravenous colistin administration was defined as a period of uninterrupted colistin administration for at least 72 hours or nine doses of continuous treatment in the same patient. Colistin formulation consisted of 150 mg colistimethate sodium (Colimycin; Kocak Farma, Istanbul, Turkey) per vial (equivalent to 1.875 million IU). All patients in the study received colistin in doses ranging 2.5 to 5 mg/kg/d in three doses, infused intravenously in 5 mL normal saline.
Statistics To test whether the data are normally distributed, the Kolmogorov-Smirnov and Shapiro Wilk tests were used. According to results, nonparametric tests were preferred. To understand if the distributions of group factors categories were homogenous among categories of nominal variables, chi-square or Fisher exact tests were used. Friedman two-way analysis of variance by ranks was used to understand, if changing by time for creatinine levels was meaningful. When the p-value from the Friedman test statistics was statistically significant, multiple comparison test was used to know which time point differ from which others. General descriptive statistics were summarized as median (minimum–maximum) for continuous variables and percentages for categorical variables. A “p” value of less than 0.05 was considered statistically significant and SPSS 15 for Windows (Chicago, IL) were used for all these statistical analyses.
Results
Definitions
Demographical, Microbiological, and Clinical Data
Standard definition for nosocomial infections and clinical sepsis were used according to the Center for Disease Control
During the study period, there were 656 admissions in the NICU, among whom 32 (5%) newborns received intravenous
American Journal of Perinatology
Vol. 31
No. 12/2014
Downloaded by: Collections and Technical Services Department. Copyrighted material.
1080
Fig. 1 Flow chart of the colistin use in our neonatal intensive care unit during study period.
colistin therapy. Eleven patients who had shorter duration of colistin treatment than 72 hours or who were born 37 weeks of gestation were excluded. The remaining 21 preterm patients were included in the study (►Fig. 1). The median GA and BW were 28 weeks (range, 23–36 weeks) and 870 g (range, 620–2,650 g), respectively. Fourteen of the 21 patients (67%) were extremely low birth weight (ELBW) preterm infants. The characteristics of preterm pa-
Alan et al.
tients who received intravenous colistin are presented in ►Table 1. All colistin treatments were started for nosocomial infection with a median postnatal age of 8 days (3–36) of life. The median initial admission weight was 915 g (555–2,610). The median daily dose of colistin was 3 mg/kg/d (2–5). Median duration and cumulative administered dose of colistin were 9 days (3–26) and 21 mg/kg (14–99), respectively. Treatment with colistin was discontinued after 3rd day of therapy in three patients due to death. Positive cultures with A. baumannii were found in 13 infants (62%) of whom 5 showed growth in peripheral blood samples, 4 in central venous catheter samples, 3 in endotracheal aspirates, and 1 in peritoneal fluid. Susceptibility testing revealed sensitivity to colistin and resistance to carbapenem, amikacin, third generation cephalosporins, and quinolones for all of 13 infants. Sensitivity to tigecycline was found only in 3 of 13 infants. Remaining eight infants received colistin empirically without any positive culture but signs of systemic infection in addition to positive acute phase reactants and no response to conventional antimicrobial treatment including carbapenem. The mortality rate was 4/21 (19%). Colistin was administered concomitantly with at least one other antimicrobial agent in all infants (►Table 2). Among 13 culture positive preterm infants, 9 (69%) showed microbiological clearance. Positive cultures with A. baumannii continued in other four patients, three of whom died during the treatment and had positive postmortem cultures with A. baumannii. One patient who had recurrent tracheal positive cultures during intravenous colistin treatment showed clinical improvement after
Table 1 Prenatal and postnatal characteristics of preterm patients receiving intravenous colistin Number of patients, n
21
Antenatal steroid, % (n)
52% (11)
Prenatal features, % (n) Maternal preeclampsia or eclampsia or essential hypertension
14% (3)
Premature prolonged ruptures of membrane or chorioamnionitis
14% (3)
Other (cervical insufficiency, maternal thyroid disorders, maternal DIC, elevated liver enzyme, polyhydramnios)
33% (7)
No prenatal problem
38% (8)
Gestational age (wk), median (range)
28 (23–36)
Birth weight (g), median (range)
870 (620–2,650)
Extremely low birth weight rate, % (n)
67% (14)
Male sex, % (n)
52% (11)
Born by Cesarean section, % (n)
76% (16)
Small for gestational age, % (n)
29% (6)
Early neonatal sepsis, % (n)
67% (14)
Number of nosocomial sepsis, median (range)
1 (1–5)
Length of stay in hospital (d), median (range)
41 (6–79)
Abbreviation: DIC, disseminated intravascular coagulation American Journal of Perinatology
Vol. 31
No. 12/2014
1081
Downloaded by: Collections and Technical Services Department. Copyrighted material.
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Alan et al.
Table 2 Concomitant antibiotics and antifungal treatment during colistin administration Patients, n (%)
Antibiotics-antifungal Meropenem
15 (71%)
Vancomicin
14 (67%)
Amikacin
8 (38%)
Amphotericin B
5 (24%)
Cefaperazone-sulbactam
3 (14%)
Ciprofloxacin
2 (9.5%)
Piperacillin-tazobactam
2 (9.5%)
Netilmicin
1 (5%)
Ampicilin-sulbactam
1 (5%)
Metronidazole
1 (5%)
Linezolide
1 (5%)
Fig. 2 Change in creatinine level showed significant increase on 6th day of colistin therapy and turned to normal levels at the end of the treatment.
aerosolized colistin therapy. A patient who had recurrent positive A. baumannii cultures during 14 days with colistin therapy could achieve a negative culture with clinical improvement after prolonged colistin therapy up to 26 days. The total recovery rate with colistin for nosocomial sepsis with/ without positive culture was found as 81% (17/21). Nine of 13 infants who had positive A. baumannii culture were ELBW preterm infants. Microbiological clearance was found in only 67% of ELBW preterm infants (6/9 infants). All three patients who could not have microbiological clearance died during the treatment course, and constituted the majority of deaths (75%, three-fourths of the infants).
Adverse Events of Colistin Administration and Outcomes The presence of renal impairment was evaluated by the AKIN criteria and it was determined in four patients (19%), three in stage 1 and one in stage 2. None of the patients required renal replacement therapy during colistin therapy or up till discharge. There was no difference in clinical comorbidities and concomitant medications which might
be related with renal impairment between AKI and non-AKI patients (p > 0.05). Patients were also evaluated for change in renal function tests and serum electrolytes during the colistin treatment. Although, creatinine level showed a significant increase and reached to a maximum level (0.92 0.42 mg/dL) on the 6th (range, 2nd–16th) day of colistin therapy (p ¼ 0.02), the change between the beginning and the end of the treatment was insignificant (p > 0.99) (►Fig. 2). Other renal function tests and mean serum electrolytes except Mg were not different before and after the colistin therapy (►Table 3). The mean serum Mg level was significantly lower at the end of the colistin therapy (1.56 0.45 mg/dL) than the beginning (1.98 0.39 mg/dL) (p < 0.001). The lowest mean Mg level was determined on the 6th day of therapy (range, 2nd–26th). When patients were evaluated for electrolyte supplementation, it was observed that at least 24% of them (24–52%) required additional electrolytes (►Table 3). Electrolyte disturbances including Mg and K were reversible in all of the surviving patients. The patient who had the longest course of colistin as 26 days in our series developed acquired
Table 3 Change in BUN, creatinine, and serum electrolytes by colistin administration Precolistin value (n ¼ 21)
Postcolistin value (n ¼ 21)
p-Value
BUN (mg/dL)
24.5 14.9
19.3 12
0.12
Creatinine (mg/dL)
0.75 0.37
0.73 0.48
0.77
Na (mEq/L)
139 (123–149)
138 (125–145)
0.15
5 (5/21, 24%)
K (mEq/L)
4.4 0.98
4.04 1.01
0.24
11 (11/21, 52%)
Ca (mg/dL)
8.53 1.26
9.18 0.97
0.10
7 (7/21, 33%)
P (mg/dL)
4.22 1.29
4.24 1.52
0.96
7 (7/21, 33%)
Mg (mg/dL)
1.98 0.39
1.56 0.45
< 0.001
11 (11/21, 52%)
Abbreviation: BUN, blood urea nitrogen. Note: Data presented as n (%), mean SD and median (range). American Journal of Perinatology
Vol. 31
No. 12/2014
Need for electrolytes supplementation during colistin therapy, n (%)
Downloaded by: Collections and Technical Services Department. Copyrighted material.
1082
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Alan et al.
1083
Table 4 Change in liver function tests by colistin treatment Precolistin value (n ¼ 21)
Postcolistin value (n ¼ 21)
p-Value
ALT (U/L)
14 (1–178)
12 (0–411)
> 0.99
AST (U/L)
39 (12–1,737)
36 (13–297)
0.68
GGT (U/L)
80 (35–376)
87 (17–366)
0.76
Total bilirubin (mg/dL)
4.04 (0.63–9.48)
3.02 (0.17–13)
0.82
Conjugated bilirubin (mg/dL)
0.52 (0.09–6.19)
0.69 (0.03–6.41)
0.18
Bartter-like syndrome during colistin therapy which has been previously published by our group.12 Liver function tests did not significantly change during colistin administration (►Table 4). No short-term neurotoxicity was observed in our group. Although, apnea was found in seven of the infants (33%) during colistin therapy, they also had other reasons of apnea (sepsis, septic shock, and/or respiratory pathologies) during the course of colistin. Two patients demonstrated delayed latency in VEP, one of whom had the longest duration of colistin treatment with 26 days. The other patient had additional comorbidity as leucomalacia which might be responsible for prolonged VEP.
Discussion Our study evaluated the effectiveness and safety of colistin in 21 preterm patients who received colistin therapy due to proven or suspected nosocomial sepsis by A. baumannii in the NICU. Our NICU is a level III referral hospital for Central Anatolia and admits especially ELBW preterm infants from all regions in contrast to low nurse to patient ratio, which may explain the high rate of MDR A. baumannii nosocomial infection. This is the largest cohort which we believe would help to increase our information on the current use of colistin in preterm infants. Our data demonstrated that colistin therapy had a high rate of efficacy up to 80% in nosocomial sepsis with A. baumannii in preterm infants. Majority of patients who did not respond to treatment were ELBW in our series. Many other studies showed favorable clinical outcome in 72 to 98% of colistin courses in children.6,8,13–15 There are only two studies which included preterm infants.7,16 Jajoo et al7 demonstrated favorable clinical outcome in 76% of colistin courses in newborns who included only eight preterm infants. In contrast, Celik et al16 reported a higher mortality rate up to 38% with preterm infants who had MDR A. baumannii infection, but only 2 of 21 patients had colistin therapy in their series. The optimal dose of colistin administration is unknown in newborn and even in children.17,18 A recent large multicenter study showed that most children were prescribed colistin according to the manufacturer’s recommendation; 2.5 to 5 mg/kg/d.5 However, some studies suggest that these doses of colistin are suboptimal and leading to delays in achieving appropriate target drug concentration.17–21 In addition, Pla-
chouras et al21 suggested that patients may benefit from a colistin loading dose due to insufficient plasma colistin concentrations before steady state. We did not use a loading dose similar to previous two studies including newborns. We suggest that the effect of loading dose can be a new research field in pharmacokinetics in newborns which may influence the rates of microbiological clearance and death. Carbapenems and/or aminoglycosides can show synergic effect with colistin to A. baumannii infection.22 In the presented study, 71% of the patients had carbapenem and 38% of the patients had amikacin as a concomitant antimicrobial therapy, but culture results showed complete in vitro resistance to these agents. On the other hand, Celik et al16 reported sensitivity to amikacin, gentamicin, colistin, imipenem, and tigecycline in their series. Only three of our patients had tigecycline sensitivity in our series but there is not yet sufficient information on use of this agent in newborns. Nephrotoxicity is the most common adverse effect of intravenous administration of colistin because the drug is excreted primarily by the kidneys.3 Although, earlier literature suggested nephrotoxicity rates approaching 50%, more recent data have reported lower nephrotoxicity rates, ranging from 0 to 30%.8,9,23–25 Multiple factors that can contribute to renal insufficiency are often present in preterm patients, including immature function of the kidney, clinical comorbidities, and concomitant nephrotoxic medication. Additional nephrotoxic agents should especially attract caution during antimicrobial management of these patients. There is only one case series of newborn patients including eight preterm infant and one case report about the effect of intravenous colistin use in preterm infants.6,12 In the study by Jajoo et al,7 only two infants developed renal impairment (> 0.5 mg/dL rise in creatinine above baseline) both of whom had severe sepsis with multiple organ dysfunction before the initiation of colistin, and one also received netilmicin concomitant with colistin. Our data demonstrate that critically ill preterm infants have had a renal impairment rate of 19% determined by AKIN criteria. Although, the rate was high this renal side effect was reversible in all surviving patients without any need of renal replacement treatment. Our series showed that many patients needed additional electrolyte supplementation during colistin therapy. Despite the intravenous magnesium supplementation, mean American Journal of Perinatology
Vol. 31
No. 12/2014
Downloaded by: Collections and Technical Services Department. Copyrighted material.
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gama glutamil transferase. Note: Data presented as n (%).
Efficacy and Safety of Intravenous Colistin in Preterm Infants serum Mg level was significantly lower at the end of the treatment. Although, there was not a significantly decrease in the mean serum potassium level at the end of colistin therapy, 52% of the patients needed additional potassium supplementation. Little information is available on the mechanism of renal toxicity but in vitro physiological studies demonstrate that, at long exposure times, colistin is directly toxic to mammalian urothelium by increasing transepithelial conduction.26 Hematuria, proteinuria, cylindruria, and renal tubular cells in urine have been reported in children.8 A recent case report by our group presented a preterm infant who had acquired Bartter-like syndrome associated with prolonged use of intravenous colistin. The patient had weight loss, polyuria, and laboratory tests revealed hypokalemic metabolic alkalosis with hypocalcemia and hypomagnesemia.12 To our knowledge, the present study is the first to address serum electrolytes for understanding renal tubulopathy associated with intravenous colistin in preterm patients. The interaction of colistin with neurons, which have high lipid content, has been associated with the occurrence of peripheral and orofacial paresthesias, visual disturbances, vertigo, mental confusion, ataxia, and seizures.27 The determination of all of these signs or symptoms is very difficult in premature infants because of immaturity. Therefore, we investigated apnea and seizure as short-term effects and VEP, BAER, and neurological deficits as long-term neurotoxicity signs. Although, apnea was found in 33% of the patients, all of them had other clinical conditions potentially related to apnea. Only two patients showed delayed VEP latency in our series, and one of them was a patient with long term use of colistin. Our study had few limitations. First, the study was designed retrospectively with lack of a control group. Second, the concomitant use of other drugs in addition to colistin and concomitant clinical conditions created difficulty in evaluation of side effects solely to colistin. In conclusion, we suggest that renal function tests and serum electrolytes should be monitored closely and replaced in case of any need during the colistin therapy in preterm infants. Although, hepatotoxicity and neurotoxicity do not appear as serious problems for preterm infants, prospective studies are needed to explore the drug’s safety in preterm infants.
References
5 Tamma PD, Newland JG, Pannaraj PS, et al. The use of intravenous
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
1 Isturiz R. Global resistance trends and the potential impact on
empirical therapy. Int J Antimicrob Agents 2008;32(Suppl 4): S201–S206 2 Giske CG, Monnet DL, Cars O, Carmeli Y; ReAct-Action on Antibiotic Resistance. Clinical and economic impact of common multidrugresistant gram-negative bacilli. Antimicrob Agents Chemother 2008;52(3):813–821 3 Nation RL, Li J. Colistin in the 21st century. Curr Opin Infect Dis 2009;22(6):535–543 4 Berlana D, Llop JM, Fort E, Badia MB, Jódar R. Use of colistin in the treatment of multiple-drug-resistant gram-negative infections. Am J Health Syst Pharm 2005;62(1):39–47
American Journal of Perinatology
Vol. 31
No. 12/2014
Alan et al.
22
23
colistin among children in the United States: results from a multicenter, case series. Pediatr Infect Dis J 2013;32(1):17–22 Iosifidis E, Antachopoulos C, Ioannidou M, et al. Colistin administration to pediatric and neonatal patients. Eur J Pediatr 2010; 169(7):867–874 Jajoo M, Kumar V, Jain M, Kumari S, Manchanda V. Intravenous colistin administration in neonates. Pediatr Infect Dis J 2011; 30(3):218–221 Falagas ME, Vouloumanou EK, Rafailidis PI. Systemic colistin use in children without cystic fibrosis: a systematic review of the literature. Int J Antimicrob Agents 2009;33(6):e1, e13 Gounden R, Bamford C, van Zyl-Smit R, Cohen K, Maartens G. Safety and effectiveness of colistin compared with tobramycin for multi-drug resistant Acinetobacter baumannii infections. BMC Infect Dis 2009;9:26 Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36(5):309–332 Mehta RL, Kellum JA, Shah SV, et al; Acute Kidney Injury Network. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11(2):R31 Cakir U, Alan S, Zeybek C, et al. Acquired bartter-like syndrome associated with colistin use in a preterm infant. Ren Fail 2013; 35(3):411–413 Celebi S, Hacımustafaoglu M, Koksal N, Ozkan H, Cetinkaya M. Colistimethate sodium therapy for multidrug-resistant isolates in pediatric patients. Pediatr Int 2010;52(3):410–414 Rosanova M, Epelbaum C, Noman A, et al. Use of colistin in a pediatric burn unit in Argentina. J Burn Care Res 2009;30(4): 612–615 Kapoor K, Jajoo M, Dublish S, Dabas V, Gupta S, Manchanda V. Intravenous colistin for multidrug-resistant gram-negative infections in critically ill pediatric patients. Pediatr Crit Care Med 2013; 14(6):e268–e272 Celik IH, Demirel G, Tatar Aksoy H, et al. Acinetobacter baumannii: an important pathogen with multidrug resistance in newborns [in Turkish]. Mikrobiyol Bul 2011;45(4):716–722 Bergen PJ, Li J, Nation RL. Dosing of colistin-back to basic PK/PD. Curr Opin Pharmacol 2011;11(5):464–469 Garonzik SM, Li J, Thamlikitkul V, 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(7):3284–3294 Markou N, Markantonis SL, Dimitrakis E, et al. Colistin serum concentrations after intravenous administration in critically ill patients with serious multidrug-resistant, gram-negative bacilli infections: a prospective, open-label, uncontrolled study. Clin Ther 2008;30(1):143–151 Daikos GL, Skiada A, Pavleas J, et al. Serum bactericidal activity of three different dosing regimens of colistin with implications for optimum clinical use. J Chemother 2010;22(3):175–178 Plachouras D, Karvanen M, Friberg LE, et al. Population pharmacokinetic analysis of colistin methanesulfonate and colistin after intravenous administration in critically ill patients with infections caused by gram-negative bacteria. Antimicrob Agents Chemother 2009;53(8):3430–3436 Petrosillo N, Ioannidou E, Falagas ME. Colistin monotherapy vs. combination therapy: evidence from microbiological, animal and clinical studies. Clin Microbiol Infect 2008;14(9): 816–827 Cheng CY, Sheng WH, Wang JT, Chen YC, Chang SC. Safety and efficacy of intravenous colistin (colistin methanesulphonate) for severe multidrug-resistant Gram-negative bacterial infections. Int J Antimicrob Agents 2010;35(3):297–300
Downloaded by: Collections and Technical Services Department. Copyrighted material.
1084
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Alan et al.
26 Lewis JR, Lewis SA. Colistin interactions with the mammalian
review of the evidence from old and recent studies. Crit Care 2006; 10(1):R27 25 Falagas ME, Sideri G, Vouloumanou EK, Papadatos JH, Kafetzis DA. Intravenous colistimethate (colistin) use in critically ill children without cystic fibrosis. Pediatr Infect Dis J 2009;28(2):123–127
urothelium. Am J Physiol Cell Physiol 2004;286(4): C913–C922 27 Spapen H, Jacobs R, Van Gorp V, Troubleyn J, Honoré PM. Renal and neurological side effects of colistin in critically ill patients. Ann Intensive Care 2011;1(1):14
Downloaded by: Collections and Technical Services Department. Copyrighted material.
24 Falagas ME, Kasiakou SK. Toxicity of polymyxins: a systematic
1085
American Journal of Perinatology
Vol. 31
No. 12/2014
1079
Efficacy and Safety of Intravenous Colistin in Preterm Infants with Nosocomial Sepsis Caused by Acinetobacter baumannii Serdar Alan, MD1 Emel Okulu, MD1
Duran Yildiz, MD1 Omer Erdeve, MD1 Ufuk Cakir, MD1 Dilek Kahvecioglu, MD1 Can Ates, MSc2 Begum Atasay, MD1 Saadet Arsan, MD1
1 Division of Neonatology, Department of Pediatrics, Ankara University
School of Medicine, Ankara, Turkey 2 Depatrment of Biostatistics, Ankara University School of Medicine, Ankara, Turkey
Address for correspondence Serdar Alan, MD, Division of Neonatology, Department of Pediatrics, Ankara University School of Medicine, Vedat Dalokay Cad. No: 90 A/16 Gaziosmanpasa, 06100 Ankara, Turkey (e-mail: [email protected]).
Abstract
Keywords
► Acinetobacter baumannii ► colistin ► neonatal intensive care unit ► nosocomial infection ► preterm
Objectives To describe the efficacy of intravenous colistin on clinical and microbiological outcomes in preterm infants with nosocomial sepsis in neonatal intensive care unit (NICU) and define adverse events observed with this treatment. Methods The records of preterm infants who received colistin with or without positive cultures in the NICU were retrospectively reviewed. Patients were evaluated for response to therapy and side effects. Results A total of 21 preterm infants with medians of 28 weeks (23–36) gestational age and 870 g (620–2,650) birth weight were included. The median duration and dose of colistin therapy were 9 days (3–26) and 3 mg/kg/d (2–5). Recovery rate in patients including all with/without positive culture was 81% (17/21). Microbiological clearance by colistin was 69% (9/13). The major side effect observed was acute kidney injury (19%). At least 24% of infants required electrolyte supplementation during the colistin therapy. Magnesium levels were significantly lower at the end of the colistin therapy (p < 0.001). Acute kidney injury and electrolyte disturbances including hypomagnesemia were reversible in all surviving patients. Conclusion We suggest that renal function tests and serum electrolytes should be monitored closely and replaced in case of any need during the colistin therapy in preterm infants.
The emergence of multidrug-resistant (MDR) nosocomial gram-negative bacteria such as Acinetobacter baumannii (A. baumannii) has become a major problem with increasing morbidity and mortality in critically ill patients including preterm infants. Meanwhile, no new antimicrobial drug to treat MDR gram-negative bacterial infection has been presented recently.1–3 Colistin which has been in use over 60 years was one of the first antibiotics with significant activity against gram-negative bacteria. However, it was largely replaced by aminoglycosides in 1970s because of
the concern about its nephrotoxicity and neurotoxicity.3 The worldwide emergence of MDR nosocomial gram-negative pathogens over the last two decades has resulted in increased use of colistin as a “salvage” therapy.4 Nephrotoxicity and neurotoxicity were reported in approximately 22 and 4% of patients, respectively, in a recent multicenter large series study in pediatric patients who were exposed to colistin for more than 72 hours.5 In contrast, two recently published trials in newborns suggested that nephrotoxicity and neurotoxicity associated with colistin were less
received January 2, 2014 accepted after revision January 21, 2014 published online February 28, 2014
Copyright © 2014 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1371361. ISSN 0735-1631.
Downloaded by: Collections and Technical Services Department. Copyrighted material.
Am J Perinatol 2014;31:1079–1086.
Efficacy and Safety of Intravenous Colistin in Preterm Infants frequent.6,7 However, there is still only scant information on intravenous colistin use especially in preterm patients.6–9 Furthermore, optimal dose and interval for safety and efficacy remain unclear. Therefore, we described the effectiveness and safety of toxicity of intravenous colistin therapy in preterm infants with and without MDR A. baumannii infection during the last 20 months.
Methods Study Design and Settings This was a retrospective study of critically ill preterm infants with proven or suspected nosocomial infection caused by MDR A. baumannii and treated with intravenous colistin. The study was conducted at a level III neonatal intensive care unit (NICU) with a capacity of 20 incubators which took care of high-risk infants of 2,500 annual inborn deliveries and highrisk referrals from all parts of Turkey. Approximately 400 patients per year were admitted to the unit. The trial was approved by the Human Research Ethics Committee of Ankara University School of Medicine.
Patient Identification and Inclusion/Exclusion Criteria All preterm patients hospitalized between January 2012 and August 2013 in the NICU and treated with intravenous colistin were evaluated. Preterm infants born before 37 weeks of gestation and treated with intravenous colistin for at least 72 hours were included in the study. Patients with major congenital anomalies were excluded.
Data Collection The medical records of all cases were reviewed from hospital files. Recorded variables included: demographic characteristics (prenatal features, gestational age [GA], birth weight [BW], sex, mode of delivery), hospital stay, postnatal age, and weight when colistin started, duration of colistin therapy, cumulative dose, concomitant drugs, and clinical condition. Blood, urine, cerebrospinal fluid, and other sterile body fluid (peritoneal fluid, etc.) cultures, and antimicrobial sensitivity were all recorded. Microbiological clearance and clinical outcomes after colistin therapy were evaluated. In addition, renal (creatinine and blood urea nitrogen level), and liver (alanine aminotransferase, aspartate aminotransferase (AST), gama glutamil transferase, and bilirubin levels) function tests and serum electrolytes (calcium, magnesium [Mg], potassium, sodium) were recorded for possible side effects before, during, and after the colistin therapy. Use of additional nephrotoxic agents that included amphotericin, aminoglycosides, diuretics, vancomycin, nonsteroidal anti-inflammatory medications, and intravenous contrast were also recorded. Short-term neurological outcomes including apnea and seizure, and long-term neurological outcomes including visual evoked potential (VEP), brainstem auditory evoked response (BAER), and neuromotor deficits were collected from patient files.
Alan et al.
and Prevention.10 MDR was defined as resistance to at least three antimicrobial agents (of different groups) with intrinsic activity against gram-negative pathogen. A favorable outcome was defined as culture negativity in culture-positive infections or clinical response with normalized acute phase reactants in clinical septic patients. The AKIN (Acute Kidney Injury Network) criteria11 were employed as a validated tool to evaluate acute kidney injury (AKI). Other clinical conditions related to AKI were defined as symptomatic patent ductus arteriosus, hypoxic ischemic birth, septic shock, hypovolemic shock, due to major bleeding, major surgery, and concomitant antimicrobial agents, including vancomycin, aminoglycosides, amphotericin B were all checked from patient files.
Microbiologic Methods BACTEC 9240 (Becton-Dickinson, NJ) was used to identify the microorganism. The VITEK 2 susceptibility card AST-N090 (bioMérieux, Marcy l’Etoile, France) containing a colistin susceptibility test was used according to the manufacturer instructions. Interpretive breakpoints (minimum inhibitory concentration [MIC] 0.5 μg/mL, susceptible, and MIC 16 μg/mL, resistant) were used for the VITEK 2.
Pharmacodynamics A course of intravenous colistin administration was defined as a period of uninterrupted colistin administration for at least 72 hours or nine doses of continuous treatment in the same patient. Colistin formulation consisted of 150 mg colistimethate sodium (Colimycin; Kocak Farma, Istanbul, Turkey) per vial (equivalent to 1.875 million IU). All patients in the study received colistin in doses ranging 2.5 to 5 mg/kg/d in three doses, infused intravenously in 5 mL normal saline.
Statistics To test whether the data are normally distributed, the Kolmogorov-Smirnov and Shapiro Wilk tests were used. According to results, nonparametric tests were preferred. To understand if the distributions of group factors categories were homogenous among categories of nominal variables, chi-square or Fisher exact tests were used. Friedman two-way analysis of variance by ranks was used to understand, if changing by time for creatinine levels was meaningful. When the p-value from the Friedman test statistics was statistically significant, multiple comparison test was used to know which time point differ from which others. General descriptive statistics were summarized as median (minimum–maximum) for continuous variables and percentages for categorical variables. A “p” value of less than 0.05 was considered statistically significant and SPSS 15 for Windows (Chicago, IL) were used for all these statistical analyses.
Results
Definitions
Demographical, Microbiological, and Clinical Data
Standard definition for nosocomial infections and clinical sepsis were used according to the Center for Disease Control
During the study period, there were 656 admissions in the NICU, among whom 32 (5%) newborns received intravenous
American Journal of Perinatology
Vol. 31
No. 12/2014
Downloaded by: Collections and Technical Services Department. Copyrighted material.
1080
Fig. 1 Flow chart of the colistin use in our neonatal intensive care unit during study period.
colistin therapy. Eleven patients who had shorter duration of colistin treatment than 72 hours or who were born 37 weeks of gestation were excluded. The remaining 21 preterm patients were included in the study (►Fig. 1). The median GA and BW were 28 weeks (range, 23–36 weeks) and 870 g (range, 620–2,650 g), respectively. Fourteen of the 21 patients (67%) were extremely low birth weight (ELBW) preterm infants. The characteristics of preterm pa-
Alan et al.
tients who received intravenous colistin are presented in ►Table 1. All colistin treatments were started for nosocomial infection with a median postnatal age of 8 days (3–36) of life. The median initial admission weight was 915 g (555–2,610). The median daily dose of colistin was 3 mg/kg/d (2–5). Median duration and cumulative administered dose of colistin were 9 days (3–26) and 21 mg/kg (14–99), respectively. Treatment with colistin was discontinued after 3rd day of therapy in three patients due to death. Positive cultures with A. baumannii were found in 13 infants (62%) of whom 5 showed growth in peripheral blood samples, 4 in central venous catheter samples, 3 in endotracheal aspirates, and 1 in peritoneal fluid. Susceptibility testing revealed sensitivity to colistin and resistance to carbapenem, amikacin, third generation cephalosporins, and quinolones for all of 13 infants. Sensitivity to tigecycline was found only in 3 of 13 infants. Remaining eight infants received colistin empirically without any positive culture but signs of systemic infection in addition to positive acute phase reactants and no response to conventional antimicrobial treatment including carbapenem. The mortality rate was 4/21 (19%). Colistin was administered concomitantly with at least one other antimicrobial agent in all infants (►Table 2). Among 13 culture positive preterm infants, 9 (69%) showed microbiological clearance. Positive cultures with A. baumannii continued in other four patients, three of whom died during the treatment and had positive postmortem cultures with A. baumannii. One patient who had recurrent tracheal positive cultures during intravenous colistin treatment showed clinical improvement after
Table 1 Prenatal and postnatal characteristics of preterm patients receiving intravenous colistin Number of patients, n
21
Antenatal steroid, % (n)
52% (11)
Prenatal features, % (n) Maternal preeclampsia or eclampsia or essential hypertension
14% (3)
Premature prolonged ruptures of membrane or chorioamnionitis
14% (3)
Other (cervical insufficiency, maternal thyroid disorders, maternal DIC, elevated liver enzyme, polyhydramnios)
33% (7)
No prenatal problem
38% (8)
Gestational age (wk), median (range)
28 (23–36)
Birth weight (g), median (range)
870 (620–2,650)
Extremely low birth weight rate, % (n)
67% (14)
Male sex, % (n)
52% (11)
Born by Cesarean section, % (n)
76% (16)
Small for gestational age, % (n)
29% (6)
Early neonatal sepsis, % (n)
67% (14)
Number of nosocomial sepsis, median (range)
1 (1–5)
Length of stay in hospital (d), median (range)
41 (6–79)
Abbreviation: DIC, disseminated intravascular coagulation American Journal of Perinatology
Vol. 31
No. 12/2014
1081
Downloaded by: Collections and Technical Services Department. Copyrighted material.
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Alan et al.
Table 2 Concomitant antibiotics and antifungal treatment during colistin administration Patients, n (%)
Antibiotics-antifungal Meropenem
15 (71%)
Vancomicin
14 (67%)
Amikacin
8 (38%)
Amphotericin B
5 (24%)
Cefaperazone-sulbactam
3 (14%)
Ciprofloxacin
2 (9.5%)
Piperacillin-tazobactam
2 (9.5%)
Netilmicin
1 (5%)
Ampicilin-sulbactam
1 (5%)
Metronidazole
1 (5%)
Linezolide
1 (5%)
Fig. 2 Change in creatinine level showed significant increase on 6th day of colistin therapy and turned to normal levels at the end of the treatment.
aerosolized colistin therapy. A patient who had recurrent positive A. baumannii cultures during 14 days with colistin therapy could achieve a negative culture with clinical improvement after prolonged colistin therapy up to 26 days. The total recovery rate with colistin for nosocomial sepsis with/ without positive culture was found as 81% (17/21). Nine of 13 infants who had positive A. baumannii culture were ELBW preterm infants. Microbiological clearance was found in only 67% of ELBW preterm infants (6/9 infants). All three patients who could not have microbiological clearance died during the treatment course, and constituted the majority of deaths (75%, three-fourths of the infants).
Adverse Events of Colistin Administration and Outcomes The presence of renal impairment was evaluated by the AKIN criteria and it was determined in four patients (19%), three in stage 1 and one in stage 2. None of the patients required renal replacement therapy during colistin therapy or up till discharge. There was no difference in clinical comorbidities and concomitant medications which might
be related with renal impairment between AKI and non-AKI patients (p > 0.05). Patients were also evaluated for change in renal function tests and serum electrolytes during the colistin treatment. Although, creatinine level showed a significant increase and reached to a maximum level (0.92 0.42 mg/dL) on the 6th (range, 2nd–16th) day of colistin therapy (p ¼ 0.02), the change between the beginning and the end of the treatment was insignificant (p > 0.99) (►Fig. 2). Other renal function tests and mean serum electrolytes except Mg were not different before and after the colistin therapy (►Table 3). The mean serum Mg level was significantly lower at the end of the colistin therapy (1.56 0.45 mg/dL) than the beginning (1.98 0.39 mg/dL) (p < 0.001). The lowest mean Mg level was determined on the 6th day of therapy (range, 2nd–26th). When patients were evaluated for electrolyte supplementation, it was observed that at least 24% of them (24–52%) required additional electrolytes (►Table 3). Electrolyte disturbances including Mg and K were reversible in all of the surviving patients. The patient who had the longest course of colistin as 26 days in our series developed acquired
Table 3 Change in BUN, creatinine, and serum electrolytes by colistin administration Precolistin value (n ¼ 21)
Postcolistin value (n ¼ 21)
p-Value
BUN (mg/dL)
24.5 14.9
19.3 12
0.12
Creatinine (mg/dL)
0.75 0.37
0.73 0.48
0.77
Na (mEq/L)
139 (123–149)
138 (125–145)
0.15
5 (5/21, 24%)
K (mEq/L)
4.4 0.98
4.04 1.01
0.24
11 (11/21, 52%)
Ca (mg/dL)
8.53 1.26
9.18 0.97
0.10
7 (7/21, 33%)
P (mg/dL)
4.22 1.29
4.24 1.52
0.96
7 (7/21, 33%)
Mg (mg/dL)
1.98 0.39
1.56 0.45
< 0.001
11 (11/21, 52%)
Abbreviation: BUN, blood urea nitrogen. Note: Data presented as n (%), mean SD and median (range). American Journal of Perinatology
Vol. 31
No. 12/2014
Need for electrolytes supplementation during colistin therapy, n (%)
Downloaded by: Collections and Technical Services Department. Copyrighted material.
1082
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Alan et al.
1083
Table 4 Change in liver function tests by colistin treatment Precolistin value (n ¼ 21)
Postcolistin value (n ¼ 21)
p-Value
ALT (U/L)
14 (1–178)
12 (0–411)
> 0.99
AST (U/L)
39 (12–1,737)
36 (13–297)
0.68
GGT (U/L)
80 (35–376)
87 (17–366)
0.76
Total bilirubin (mg/dL)
4.04 (0.63–9.48)
3.02 (0.17–13)
0.82
Conjugated bilirubin (mg/dL)
0.52 (0.09–6.19)
0.69 (0.03–6.41)
0.18
Bartter-like syndrome during colistin therapy which has been previously published by our group.12 Liver function tests did not significantly change during colistin administration (►Table 4). No short-term neurotoxicity was observed in our group. Although, apnea was found in seven of the infants (33%) during colistin therapy, they also had other reasons of apnea (sepsis, septic shock, and/or respiratory pathologies) during the course of colistin. Two patients demonstrated delayed latency in VEP, one of whom had the longest duration of colistin treatment with 26 days. The other patient had additional comorbidity as leucomalacia which might be responsible for prolonged VEP.
Discussion Our study evaluated the effectiveness and safety of colistin in 21 preterm patients who received colistin therapy due to proven or suspected nosocomial sepsis by A. baumannii in the NICU. Our NICU is a level III referral hospital for Central Anatolia and admits especially ELBW preterm infants from all regions in contrast to low nurse to patient ratio, which may explain the high rate of MDR A. baumannii nosocomial infection. This is the largest cohort which we believe would help to increase our information on the current use of colistin in preterm infants. Our data demonstrated that colistin therapy had a high rate of efficacy up to 80% in nosocomial sepsis with A. baumannii in preterm infants. Majority of patients who did not respond to treatment were ELBW in our series. Many other studies showed favorable clinical outcome in 72 to 98% of colistin courses in children.6,8,13–15 There are only two studies which included preterm infants.7,16 Jajoo et al7 demonstrated favorable clinical outcome in 76% of colistin courses in newborns who included only eight preterm infants. In contrast, Celik et al16 reported a higher mortality rate up to 38% with preterm infants who had MDR A. baumannii infection, but only 2 of 21 patients had colistin therapy in their series. The optimal dose of colistin administration is unknown in newborn and even in children.17,18 A recent large multicenter study showed that most children were prescribed colistin according to the manufacturer’s recommendation; 2.5 to 5 mg/kg/d.5 However, some studies suggest that these doses of colistin are suboptimal and leading to delays in achieving appropriate target drug concentration.17–21 In addition, Pla-
chouras et al21 suggested that patients may benefit from a colistin loading dose due to insufficient plasma colistin concentrations before steady state. We did not use a loading dose similar to previous two studies including newborns. We suggest that the effect of loading dose can be a new research field in pharmacokinetics in newborns which may influence the rates of microbiological clearance and death. Carbapenems and/or aminoglycosides can show synergic effect with colistin to A. baumannii infection.22 In the presented study, 71% of the patients had carbapenem and 38% of the patients had amikacin as a concomitant antimicrobial therapy, but culture results showed complete in vitro resistance to these agents. On the other hand, Celik et al16 reported sensitivity to amikacin, gentamicin, colistin, imipenem, and tigecycline in their series. Only three of our patients had tigecycline sensitivity in our series but there is not yet sufficient information on use of this agent in newborns. Nephrotoxicity is the most common adverse effect of intravenous administration of colistin because the drug is excreted primarily by the kidneys.3 Although, earlier literature suggested nephrotoxicity rates approaching 50%, more recent data have reported lower nephrotoxicity rates, ranging from 0 to 30%.8,9,23–25 Multiple factors that can contribute to renal insufficiency are often present in preterm patients, including immature function of the kidney, clinical comorbidities, and concomitant nephrotoxic medication. Additional nephrotoxic agents should especially attract caution during antimicrobial management of these patients. There is only one case series of newborn patients including eight preterm infant and one case report about the effect of intravenous colistin use in preterm infants.6,12 In the study by Jajoo et al,7 only two infants developed renal impairment (> 0.5 mg/dL rise in creatinine above baseline) both of whom had severe sepsis with multiple organ dysfunction before the initiation of colistin, and one also received netilmicin concomitant with colistin. Our data demonstrate that critically ill preterm infants have had a renal impairment rate of 19% determined by AKIN criteria. Although, the rate was high this renal side effect was reversible in all surviving patients without any need of renal replacement treatment. Our series showed that many patients needed additional electrolyte supplementation during colistin therapy. Despite the intravenous magnesium supplementation, mean American Journal of Perinatology
Vol. 31
No. 12/2014
Downloaded by: Collections and Technical Services Department. Copyrighted material.
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gama glutamil transferase. Note: Data presented as n (%).
Efficacy and Safety of Intravenous Colistin in Preterm Infants serum Mg level was significantly lower at the end of the treatment. Although, there was not a significantly decrease in the mean serum potassium level at the end of colistin therapy, 52% of the patients needed additional potassium supplementation. Little information is available on the mechanism of renal toxicity but in vitro physiological studies demonstrate that, at long exposure times, colistin is directly toxic to mammalian urothelium by increasing transepithelial conduction.26 Hematuria, proteinuria, cylindruria, and renal tubular cells in urine have been reported in children.8 A recent case report by our group presented a preterm infant who had acquired Bartter-like syndrome associated with prolonged use of intravenous colistin. The patient had weight loss, polyuria, and laboratory tests revealed hypokalemic metabolic alkalosis with hypocalcemia and hypomagnesemia.12 To our knowledge, the present study is the first to address serum electrolytes for understanding renal tubulopathy associated with intravenous colistin in preterm patients. The interaction of colistin with neurons, which have high lipid content, has been associated with the occurrence of peripheral and orofacial paresthesias, visual disturbances, vertigo, mental confusion, ataxia, and seizures.27 The determination of all of these signs or symptoms is very difficult in premature infants because of immaturity. Therefore, we investigated apnea and seizure as short-term effects and VEP, BAER, and neurological deficits as long-term neurotoxicity signs. Although, apnea was found in 33% of the patients, all of them had other clinical conditions potentially related to apnea. Only two patients showed delayed VEP latency in our series, and one of them was a patient with long term use of colistin. Our study had few limitations. First, the study was designed retrospectively with lack of a control group. Second, the concomitant use of other drugs in addition to colistin and concomitant clinical conditions created difficulty in evaluation of side effects solely to colistin. In conclusion, we suggest that renal function tests and serum electrolytes should be monitored closely and replaced in case of any need during the colistin therapy in preterm infants. Although, hepatotoxicity and neurotoxicity do not appear as serious problems for preterm infants, prospective studies are needed to explore the drug’s safety in preterm infants.
References
5 Tamma PD, Newland JG, Pannaraj PS, et al. The use of intravenous
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
1 Isturiz R. Global resistance trends and the potential impact on
empirical therapy. Int J Antimicrob Agents 2008;32(Suppl 4): S201–S206 2 Giske CG, Monnet DL, Cars O, Carmeli Y; ReAct-Action on Antibiotic Resistance. Clinical and economic impact of common multidrugresistant gram-negative bacilli. Antimicrob Agents Chemother 2008;52(3):813–821 3 Nation RL, Li J. Colistin in the 21st century. Curr Opin Infect Dis 2009;22(6):535–543 4 Berlana D, Llop JM, Fort E, Badia MB, Jódar R. Use of colistin in the treatment of multiple-drug-resistant gram-negative infections. Am J Health Syst Pharm 2005;62(1):39–47
American Journal of Perinatology
Vol. 31
No. 12/2014
Alan et al.
22
23
colistin among children in the United States: results from a multicenter, case series. Pediatr Infect Dis J 2013;32(1):17–22 Iosifidis E, Antachopoulos C, Ioannidou M, et al. Colistin administration to pediatric and neonatal patients. Eur J Pediatr 2010; 169(7):867–874 Jajoo M, Kumar V, Jain M, Kumari S, Manchanda V. Intravenous colistin administration in neonates. Pediatr Infect Dis J 2011; 30(3):218–221 Falagas ME, Vouloumanou EK, Rafailidis PI. Systemic colistin use in children without cystic fibrosis: a systematic review of the literature. Int J Antimicrob Agents 2009;33(6):e1, e13 Gounden R, Bamford C, van Zyl-Smit R, Cohen K, Maartens G. Safety and effectiveness of colistin compared with tobramycin for multi-drug resistant Acinetobacter baumannii infections. BMC Infect Dis 2009;9:26 Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36(5):309–332 Mehta RL, Kellum JA, Shah SV, et al; Acute Kidney Injury Network. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11(2):R31 Cakir U, Alan S, Zeybek C, et al. Acquired bartter-like syndrome associated with colistin use in a preterm infant. Ren Fail 2013; 35(3):411–413 Celebi S, Hacımustafaoglu M, Koksal N, Ozkan H, Cetinkaya M. Colistimethate sodium therapy for multidrug-resistant isolates in pediatric patients. Pediatr Int 2010;52(3):410–414 Rosanova M, Epelbaum C, Noman A, et al. Use of colistin in a pediatric burn unit in Argentina. J Burn Care Res 2009;30(4): 612–615 Kapoor K, Jajoo M, Dublish S, Dabas V, Gupta S, Manchanda V. Intravenous colistin for multidrug-resistant gram-negative infections in critically ill pediatric patients. Pediatr Crit Care Med 2013; 14(6):e268–e272 Celik IH, Demirel G, Tatar Aksoy H, et al. Acinetobacter baumannii: an important pathogen with multidrug resistance in newborns [in Turkish]. Mikrobiyol Bul 2011;45(4):716–722 Bergen PJ, Li J, Nation RL. Dosing of colistin-back to basic PK/PD. Curr Opin Pharmacol 2011;11(5):464–469 Garonzik SM, Li J, Thamlikitkul V, 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(7):3284–3294 Markou N, Markantonis SL, Dimitrakis E, et al. Colistin serum concentrations after intravenous administration in critically ill patients with serious multidrug-resistant, gram-negative bacilli infections: a prospective, open-label, uncontrolled study. Clin Ther 2008;30(1):143–151 Daikos GL, Skiada A, Pavleas J, et al. Serum bactericidal activity of three different dosing regimens of colistin with implications for optimum clinical use. J Chemother 2010;22(3):175–178 Plachouras D, Karvanen M, Friberg LE, et al. Population pharmacokinetic analysis of colistin methanesulfonate and colistin after intravenous administration in critically ill patients with infections caused by gram-negative bacteria. Antimicrob Agents Chemother 2009;53(8):3430–3436 Petrosillo N, Ioannidou E, Falagas ME. Colistin monotherapy vs. combination therapy: evidence from microbiological, animal and clinical studies. Clin Microbiol Infect 2008;14(9): 816–827 Cheng CY, Sheng WH, Wang JT, Chen YC, Chang SC. Safety and efficacy of intravenous colistin (colistin methanesulphonate) for severe multidrug-resistant Gram-negative bacterial infections. Int J Antimicrob Agents 2010;35(3):297–300
Downloaded by: Collections and Technical Services Department. Copyrighted material.
1084
Efficacy and Safety of Intravenous Colistin in Preterm Infants
Alan et al.
26 Lewis JR, Lewis SA. Colistin interactions with the mammalian
review of the evidence from old and recent studies. Crit Care 2006; 10(1):R27 25 Falagas ME, Sideri G, Vouloumanou EK, Papadatos JH, Kafetzis DA. Intravenous colistimethate (colistin) use in critically ill children without cystic fibrosis. Pediatr Infect Dis J 2009;28(2):123–127
urothelium. Am J Physiol Cell Physiol 2004;286(4): C913–C922 27 Spapen H, Jacobs R, Van Gorp V, Troubleyn J, Honoré PM. Renal and neurological side effects of colistin in critically ill patients. Ann Intensive Care 2011;1(1):14
Downloaded by: Collections and Technical Services Department. Copyrighted material.
24 Falagas ME, Kasiakou SK. Toxicity of polymyxins: a systematic
1085
American Journal of Perinatology
Vol. 31
No. 12/2014
