Early administration of mucoactive agents and ventilator-free days: a propensity score-matched study
Highlight box
Key findings
• Early administration of mucoactive agents did not increase the number of ventilator-free days.
What is known and what is new?
• Although mucoactive agents are often administered in critically ill patients, the benefits of expectorant administration in mechanically ventilated patients are inconclusive.
• Administration of mucoactive agents within 3 days of admission to the intensive care unit did not increase the number of ventilator-free days in critically ill patients.
What is the implication, and what should change now?
• Further studies are needed to determine which patient populations will benefit from mucoactive agents.
Introduction
In mechanically ventilated patients, mucociliary transport and coughing, the primary mechanisms for clearance of secretions, are impaired (1). In addition, artificial airways, inadequate humidification, and immobility can result in the accumulation of secretions, leading to ventilation/perfusion mismatch and increased work of breathing, which subsequently lead to a cycle of secretion increase (1). A previous study reported that atelectasis was found in 14% of mechanically ventilated trauma patients (2). Accordingly, the management of airway secretions is a critical issue, and mucoactive agents are often prescribed as part of the management of airway secretions.
In a survey of critically ill patients in the United Kingdom, 41% of mechanically ventilated patients received at least one mucoactive agent (3). Mucoactive agents include expectorants, mucolytics, mucoregulatory, and mucokinetic drugs, which are intended to increase the ability to expectorate secretions or reduce mucus hypersecretion (4). In a randomized controlled trial (RCT) on mechanically ventilated patients, on-demand compared with routine nebulization of NAC with salbutamol did not result in fewer ventilator-free days (VFDs) (5). In another RCT (6), intravenous NAC was administered early after admission in ventilated patients with acute respiratory failure, resulting in a shorter duration of ventilation. However, these studies had small sample sizes, and there was high heterogeneity; consequently, a meta-analysis concluded that the existing evidence was of low quality (7). To date, thus, the benefits of early or routine administration of mucoactive agents in ventilated patients are inconclusive.
Therefore, we aimed to examine whether the early administration of mucoactive agents in mechanically ventilated patients is associated with an increase in VFDs. We present the following article in accordance with the STROBE reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-22-4340/rc).
Methods
Study design and patients
This retrospective observational study used electronic medical records and claims data from a Japanese hospital. The hospital is a tertiary care facility located in a rural city in western Japan. Two intensive care units (ICUs) participated in the study. One ICU had eight beds, admitting emergency patients with acute medical or surgical diseases. The other ICU had 10 beds, admitting those who underwent elective surgery and those who deteriorated after hospitalization on general wards.
We included all patients who were aged 18 years or older, received invasive mechanical ventilation, and were admitted to the two ICUs between November 1, 2018, and March 31, 2021. We excluded patients who did not receive invasive mechanical ventilation on the day of ICU admission and those who started mechanical ventilation at another ICU in the hospital or other facilities. Patients whose body mass index (BMI) data were lacking or whose BMI was less than 14 kg/m2 or more than 40 kg/m2 were excluded from the study. We also excluded patients who were discharged from the ICU within 3 days of admission.
Ethics
The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics committee of Kurashiki Central Hospital (No. 3649, June 22, 2021), and individual consent for this retrospective analysis was waived.
Variables
From the patients’ electronic medical records, we collected data on their demographics (age, sex, height, weight, comorbidities, and history of smoking), sepsis-related organ failure assessment (SOFA) score (8), norepinephrine or epinephrine administration, length of ICU and hospital stay, as well as ICU and hospital mortality. BMI was calculated as the ratio of weight (kg) to height squared (m2). We gathered data on the dates of the procedures (e.g., general anesthesia, tracheal intubation, mechanical ventilation, and tracheostomy) from the patient claims data. We also extracted the names and starting dates of the mucoactive agents prescribed to the patients; they included NAC, ambroxol, L-carbocisteine, fudosteine, and bromhexine. Because we did not use either hypertonic saline inhalation or heparin inhalation in ventilated patients in our hospital, we did not evaluate them in the present study.
We categorized the admitting diagnoses into the following eight groups according to the tenth revision of the International Statistical Classification of Diseases and Related Health Problems: infectious disease (A00–B99), neoplasms and hematological and immunological diseases (C00–D89), neurological (G00–G99), cardiovascular (I00–I99), respiratory (J00–J99), gastrointestinal and liver (K00–K93), trauma (S00–T98), and others.
Our primary outcome was VFDs during the first 28 days of ICU stay. We defined VFDs as days alive and free of mechanical ventilation in the first 28 days from intubation. We defined VFDs as 0 if the patient died within 28 days or received invasive mechanical ventilation for more than 28 days (9). Invasive mechanical ventilation was defined using patient claims data. The claims data used the Diagnostic Procedure Combination/Per Diem Payment System, a case-mix payment system for acutely hospitalized patients based on diagnoses and procedures that includes information on the details of the procedures coded with unique Japanese procedure codes and the dates the procedures were performed (10). Invasive mechanical ventilation was defined as mechanical ventilation (Japanese code J045) following general anesthesia (L008) or tracheal intubation (J044). As a secondary outcome, we used ICU-free days from the first day of ICU admission. We defined ICU-free days as days alive and free of ICU admission in the first 28 days from intubation. We defined ICU-free days as 0 if the patient died within 28 days or stayed in the ICU for more than 28 days.
Statistical analysis
For the descriptive analysis of patient characteristics and outcomes, we used Fisher’s exact test for categorical variables and the Kruskal-Wallis test for continuous variables.
We applied 1:1 propensity score matching analysis between the early mucoactive agent group (patients who had the first prescription of any mucoactive agents within 3 days after ICU admission) and the on-demand mucoactive agent group (patients who had the first prescription of mucoactive agents after the fourth day of ICU admission or patients who had no prescription during their hospitalization) to reduce selection bias. We selected prescriptions of mucoactive agents within 3 days of ICU admission as the exposure to evaluate whether early mucoactive agent administration improved VFDs. We estimated a propensity score as the probability that the patient will be prescribed mucoactive agents within 3 days of ICU admission using a multivariate logistic regression model. The model included the following factors: sex, age, chronic obstructive pulmonary disease (J43.1, J43.2, J43.8, J43.9, J44.0, J44.1, J44.8, J44.9), chronic respiratory failure (J96.1), chronic heart failure (I50.9), congestive heart failure (I50.0), history of smoking, BMI, SOFA score, and category of admitting diagnoses. The matching procedure was performed using the nearest neighbor method without replacement and with a caliper width of 0.2 of the pooled standard deviation of the logit of the propensity score. After matching, we assessed the patient background, balancing between the two groups using standardized mean differences (SMDs) and defined the covariate imbalance threshold as SMD >0.2 (11). We conducted a sensitivity analysis using a 1:4 matching algorithm between the patients in the carbocisteine early and on-demand mucoactive agent groups. We performed the subsequent procedures as described in the main analysis.
All analyses were performed using Stata version 16.1 software (Stata, College Station, TX, USA). We considered P values <0.05 (two-tailed) as statistically significant.
Results
In total, 662 patients were included in the analysis (Figure 1). Patient characteristics before matching are shown in Table 1. Sixty-eight patients (10.3%) were prescribed mucoactive agents within 3 days of ICU admission. Of those, 24 patients (35.3%) had early prescriptions of multiple mucoactive agents. The VFDs during the first 28 days of ICU stay were 21 [interquartile range (IQR): 6–24] and 21 (IQR: 13–24) for the early and on-demand mucoactive agent groups, respectively. Figure 2. shows the number of ventilator-free days for the early mucoactive agent group and the on-demand mucoactive agent group. The number of patients prescribed each mucoactive agent is shown in Table 2. Within 3 days of ICU admission, carbocisteine was the most commonly prescribed mucoactive agent (n=48, 70.6%).
Table 1
| Characteristics | Total, n=662 | On-demand mucoactive agents, n=594 | Early mucoactive agents, n=68 | P value |
|---|---|---|---|---|
| Age, year, median [IQR] | 72 [61–80] | 72 [63.5–80] | 72 [60–80] | 0.48 |
| Sex, male, n (%) | 434 (65.6) | 378 (63.6) | 56 (82.4) | 0.00 |
| SOFA score, median [IQR] | 6 [4–10] | 5 [2–7] | 7 [4–10] | 0.00 |
| BMI (kg/m2), median [IQR] | 22.97 [20.3–25.7] | 22.3 [18.7–24.7] | 23.0 [20.3–25.8] | 0.05 |
| Admitting disease group, n (%) | ||||
| Infectious disease | 42 (6.3) | 38 (6.4) | 4 (5.9) | |
| Neoplasms, hematological and immunological diseases | 77 (11.6) | 68 (11.5) | 9 (13.2) | |
| Neurological | 53 (8.0) | 44 (7.4) | 9 (13.2) | |
| Cardiovascular | 34 (5.1) | 31 (5.2) | 3 (4.4) | |
| Respiratory | 119 (18.0) | 97 (16.3) | 22 (32.4) | |
| Gastrointestinal and liver | 111 (16.8) | 105 (17.7) | 6 (8.8) | |
| Trauma | 101 (15.3) | 96 (16.2) | 5 (7.4) | |
| Other | 125 (18.9) | 115 (19.4) | 10 (14.7) | 0.01 |
| Underlying disease, n (%) | ||||
| Diabetes mellites | 83 (12.5) | 71 (12.0) | 12 (17.7) | 0.18 |
| Chronic respiratory failure | 30 (4.5) | 19 (3.2) | 11 (16.2) | 0.00 |
| Chronic heart failure | 45 (6.8) | 40 (6.7) | 5 (7.4) | 0.80 |
| Chronic kidney disease | 43 (6.5) | 42 (7.1) | 1 (1.5) | 0.11 |
| Liver cirrhosis | 22 (3.3) | 21 (3.5) | 1 (1.5) | 0.72 |
| Pneumonia on hospital admission, n (%) | 52 (7.9) | 39 (6.6) | 13 (19.1) | 0.00 |
| Smoking, n (%) | 304 (49.9) | 253 (46.3) | 51 (82.3) | 0.00 |
| Epinepheline, n (%) | 92 (13.9) | 88 (14.8) | 4 (5.9) | 0.04 |
| Norepinephrine, n (%) | 594 (89.7) | 535 (90.1) | 59 (86.8) | 0.28 |
| Hospital mortality, n (%) | 104 (15.7) | 93 (15.7) | 11 (16.2) | 0.86 |
| Ventilator free days (at 28 days), median [IQR] | 21 [13–24] | 21 [6–24] | 21 [13–24] | 0.55 |
| Length of ICU stay, median [IQR] | 7 [5–11] | 7 [5–10] | 7 [5–11] | 0.39 |
| Length of ICU stay, total, median [IQR] | 8 [5–12] | 8 [5–12.5] | 8 [5–12] | 0.91 |
| Length of hospital stay, median [IQR] | 30 [18–49] | 33 [19–47.5] | 30 [18–49] | 0.52 |
| Ventilator days, median [IQR] | 6 [4–12] | 6.5 [4–16] | 6 [4–12] | 0.41 |
IQR, interquartile range; SOFA, Sequential Organ Failure Assessment; BMI, body mass index; ICU, intensive care unit.
Table 2
| Mucoactive agents | Value, n (%) |
|---|---|
| Ambroxol | 17 (25.0) |
| Bromhexine (inhale) | 8 (11.8) |
| Bromhexine (intravenous) | 16 (23.5) |
| Carbocisteine | 48 (70.6) |
| Fudosteine | 11 (16.2) |
| N-acetylcysteine (inhale) | 0 (0.0) |
After propensity score matching, a total of 94 patients were included in the analysis (47 in each group). After matching, we found that the SMD for all covariates was <0.2, indicating that they were well balanced between the groups (Table 3). Of the patients in the on-demand mucoactive agent group, six received mucoactive agents during their ICU stay and 31 received them during their hospitalization. Table 4 shows the median VFDs: 21 (IQR: 1–24) in the early mucoactive agent group and 20 (IQR: 13–24) in the on-demand mucoactive agent group (P=0.53). We also compared ICU-free days: the median values were 19 (IQR: 12–22) and 19 (IQR: 13–22) for the early and on-demand mucoactive agent groups, respectively (P=0.72). In the sensitivity analysis, we found no difference in both VFDs [20.5 (IQR: 4.5–24) vs. 20 (IQR: 2–23), P=0.51] and ICU-free days [18.5 (IQR: 13–23) vs. 19 (IQR: 9–22), P=0.54] in the early carbocisteine group compared to the non-early mucoactive agent group (Table 4).
Table 3
| Characteristics | On-demand mucoactive agents, n=47 | Early mucoactive agents, n=47 | SMD |
|---|---|---|---|
| Age, years, median [IQR] | 75 [64–80] | 72 [62–78] | −0.139 |
| Sex, male, n (%) | 34 (72.3) | 37 (78.7) | 0.147 |
| Smoking, n (%) | 34 (72.3) | 37 (78.7) | 0.147 |
| SOFA score, median [IQR] | 4 [2–8] | 4 [2–8] | −0.042 |
| Chronic respiratory failure, n (%) | 6 (12.8) | 4 (8.5) | −0.137 |
| Chronic heart failure, n (%) | 5 (10.6) | 3 (6.4) | −0.151 |
| Smoking, n (%) | 34 (72.3) | 37 (78.7) | 0.147 |
| Admitting disease group, n (%) | |||
| Infectious disease | 3 (6.4) | 2 (4.3) | |
| Neoplasms, hematological and immunological diseases | 3 (6.4) | 6 (12.8) | |
| Neurological | 7 (14.9) | 7 (14.9) | |
| Cardiovascular | 2 (4.3) | 3 (6.4) | |
| Respiratory | 17 (36.2) | 15 (31.9) | |
| Gastrointestinal and liver | 4 (8.5) | 4 (8.5) | |
| Trauma | 7 (14.9) | 4 (8.5) | |
| Other | 4 (8.5) | 6 (12.8) | −0.065 |
IQR, interquartile range; SOFA score, Sequential Organ Failure Assessment score; SMD, standardized mean difference.
Table 4
| Outcomes | On-demand mucoactive agents, n=47 | Early mucoactive agents, n=47 | P |
|---|---|---|---|
| Ventilator-free days, median [IQR] | 21 [13–24] | 20 [1–24] | 0.534 |
| ICU-free days, median [IQR] | 19 [13–22] | 19 [12–22] | 0.724 |
ICU, intensive care unit; IQR, interquartile range.
Discussion
In this propensity analysis, we examined the association between eary administration of mucoactive agents and VFDs in critically ill patients. Compared with the on-demand mucoactive agent group, VFDs did not improve among the early mucoactive agent group, even after adjusting for potential confounders. Mucoactive agents promote airway clearance by affecting mucociliary clearance and mucus properties (4). Consequently, the use of mucoactive agents to improve airway clearance in ventilated patients seems theoretically beneficial. However, supporting evidence is limited (12), and a meta-analysis did not show the benefit of mucoactive agents in critically ill patients with acute respiratory failure (7). Similarly, our study suggested no increase in the VFDs or ICU-free days with early prescription of mucoactive agents compared with on-demand prescription of such agents.
In Japan, ambroxol is not available for intravenous use, and it was prescribed to only 25% of patients who had an early prescription of mucoactive agents in this study. In a conference proceeding, it was reported that intravenous ambroxol shortened the duration of mechanical ventilation and ICU stay by 2.0 days in an RCT of 68 patients with acute respiratory distress syndrome (13). In the present study, the smaller number of patients to whom ambroxol was prescribed, the delivery route, and the administered dose and duration differed from the previous study, which may have failed to prove any benefit. The bioavailability of ambroxol after oral administration is above 70%, which is not poor (14), but the dose in the RCT was significantly high. Carbocisteine was the most commonly used mucoactive agent in our study. Carbocisteine was used in 16% of ventilated patients in ICUs in the United Kingdom (3). Despite its widespread use, evidence on carbocisteine in ventilated patients is lacking (7). We performed a sensitivity analysis of carbocisteine and found no difference in both VFDs and ICU-free days compared with the on-demand mucoactive agent group (Table 4).
There are several strengths to this study. To our knowledge, this is the first study to compare an early vs. on-demand use of mucoactive agents in mechanically ventilated patients. Previous studies mainly compared mucoactive agents vs placebo with a meta-analysis showing no difference in mortality or duration of mechanical ventilation but an improvement in ICU length of stay (7). Our study used a clinically relevant endpoint of ventilator-free days and used propensity matching to reduce the risk of confounding in this non-randomized study. Especially given the heterogeneity of patients included in our study.
This study had some limitations. First, there was no information about whether a heated humidifier (HH) or a heat and moisture exchanger (HME) was used in the patients’ ventilator circuits. Humidification of air is routinely performed during mechanical ventilation to prevent problems such as airway obstruction and drop in body temperature (15). Humidification methods include HH, an active form of humidification, and HME, a passive form of humidification. HMEs have lower humidification levels than HHs and are associated with increased airway obstruction (16). Nevertheless, a meta-analysis found no difference in mortality, pneumonia incidence, or artificial airway occlusion (17), and the method of airway humidification had little effect on the duration of mechanical ventilation and length of ICU stay. Thus, the lack of information regarding HH or HME use might have limited impacts on our study findings. Second, the patients in the early prescribing group were included regardless of the details of their prescriptions (number or type of mucoactive agents) as long as they were within 3 days of ICU admission. However, different mucoactive agents may exert different effects. Further studies are needed for each mucoactive agent individually, to examine the impact of the number or type of mucoactive agents on airway secretions (in mechanically ventilated patients). Third, we had no information on whether the patients had been taking mucoactive agents before ICU admission, and we could not assess the effect of any preceding mucoactive drugs. Fourth, we did not adjust for the patients’ respiratory status or ventilator settings in the analysis. However, the patient’s disease, comorbidities, and severity scores at admission were used to calculate the propensity score, adjusting for indication bias. Fifth, we did not assess the incidence of ventilator-associated pneumonia (VAP). However, the international consensus on core outcome sets for critical care ventilation trials does not include VAP (18). Furthermore, VAP occurs after prolonged ventilation or following inadequate management of airway clearance in the presence of a tracheal tube (19). Accordingly, we considered VAP to be an intermediate variable and did not adjust for it in our study.
Oral or injectable mucoactive agents are inexpensive and may be associated with a reduced workload for medical staff than nebulized mucoactive agents. However, given that mucoactive agents are used in many critically ill patients, their cost is potentially significant. Therefore, studies to examine what type of patient population will benefit from mucoactive agents or whether they are effective in those who are at risk of airway secretion retention and pulmonary complications are needed. Further evidence is also required to determine the efficacy of mucoactive agents in critically ill patients with acute respiratory failure.
Conclusions
In this propensity-matched analysis, we found no significant increase in the VFDs and ICU-free days in the group with early administration of mucoactive agents. Further studies should focus on patients with airway clearance problems and not only on those with acute respiratory failure, as well as individual mucoactive agents (such as ambroxol and carbocisteine) that previous studies have not sufficiently assessed.
Acknowledgments
Funding: None.
Footnote
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://atm.amegroups.com/article/view/10.21037/atm-22-4340/rc
Data Sharing Statement: Available at https://atm.amegroups.com/article/view/10.21037/atm-22-4340/dss
Peer Review File: Available at https://atm.amegroups.com/article/view/10.21037/atm-22-4340/prf
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-22-4340/coif). JF reports the Health, Labour and Welfare Policy Research Grants (Grant Number 19IA2024). The other authors have no conflicts of interest to declare.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics committee of Kurashiki Central Hospital (No. 3649, June 22, 2021), and individual consent for this retrospective analysis was waived.
Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
References
- Volpe MS, Guimarães FS, Morais CC. Airway Clearance Techniques for Mechanically Ventilated Patients: Insights for Optimization. Respir Care 2020;65:1174-88. [Crossref] [PubMed]
- Hongrattana G, Reungjui P, Tumsatan P, et al. Incidence and risk factors of pulmonary atelectasis in mechanically ventilated trauma patients in ICU: a prospective study. Int J Evid Based Healthc 2019;17:44-52. [Crossref] [PubMed]
- Borthwick M, McAuley D, Warburton J, et al. Mucoactive agent use in adult UK Critical Care Units: a survey of health care professionals’ perception, pharmacists’ description of practice, and point prevalence of mucoactive use in invasively mechanically ventilated patients. PeerJ 2020;8:e8828. [Crossref] [PubMed]
- Rubin BK. Mucolytics, expectorants, and mucokinetic medications. Respir Care 2007;52:859-65. [PubMed]
- van Meenen DMP, van der Hoeven SM, Binnekade JM, et al. Effect of On-Demand vs Routine Nebulization of Acetylcysteine With Salbutamol on Ventilator-Free Days in Intensive Care Unit Patients Receiving Invasive Ventilation: A Randomized Clinical Trial. JAMA 2018;319:993-1001. [Crossref] [PubMed]
- Moradi M, Mojtahedzadeh M, Mandegari A, et al. The role of glutathione-S-transferase polymorphisms on clinical outcome of ALI/ARDS patient treated with N-acetylcysteine. Respir Med 2009;103:434-41. [Crossref] [PubMed]
- Anand R, McAuley DF, Blackwood B, et al. Mucoactive agents for acute respiratory failure in the critically ill: a systematic review and meta-analysis. Thorax 2020;75:623-31. [Crossref] [PubMed]
- Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707-10. [Crossref] [PubMed]
- Yehya N, Harhay MO, Curley MAQ, et al. Reappraisal of Ventilator-Free Days in Critical Care Research. Am J Respir Crit Care Med 2019;200:828-36. [Crossref] [PubMed]
- Nakamura K. Diagnosis Procedure Combination Database Would Develop Nationwide Clinical Research in Japan. Circ J 2016;80:2289-90. [Crossref] [PubMed]
- Jackson JW, Schmid I, Stuart EA. Propensity Scores in Pharmacoepidemiology: Beyond the Horizon. Curr Epidemiol Rep 2017;4:271-80. [Crossref] [PubMed]
- Strickland SL, Rubin BK, Haas CF, et al. AARC Clinical Practice Guideline: Effectiveness of Pharmacologic Airway Clearance Therapies in Hospitalized Patients. Respir Care 2015;60:1071-7. [Crossref] [PubMed]
- Yong-Jun L, Juan C, Bin O, et al. Effect of high dose ambroxol on the extra-vascular lung water and oxygenation of patients with extra-pulmonary acute respiratory distress syndrome. Intensive Care Med 2014;40:s233.
- Vergin H, Bishop-Freudling GB, Miczka M, et al. The pharmacokinetics and bioequivalence of various dosage forms of ambroxol. Arzneimittelforschung 1985;35:1591-5. [PubMed]
- Park S, Yoon SH, Youn AM, et al. Heated wire humidification circuit attenuates the decrease of core temperature during general anesthesia in patients undergoing arthroscopic hip surgery. Korean J Anesthesiol 2017;70:619-25. [Crossref] [PubMed]
- American Association for Respiratory Care. Restrepo RD, Walsh BK. Humidification during invasive and noninvasive mechanical ventilation: 2012. Respir Care 2012;57:782-8. [Crossref] [PubMed]
- Gillies D, Todd DA, Foster JP, et al. Heat and moisture exchangers versus heated humidifiers for mechanically ventilated adults and children. Cochrane Database Syst Rev 2017;9:CD004711. [Crossref] [PubMed]
- Blackwood B, Ringrow S, Clarke M, et al. A Core Outcome Set for Critical Care Ventilation Trials. Crit Care Med 2019;47:1324-31. [Crossref] [PubMed]
- Kalanuria AA, Ziai W, Mirski M. Ventilator-associated pneumonia in the ICU. Crit Care 2014;18:208. [Crossref] [PubMed]
