Journal of Anesthesia

, Volume 30, Issue 5, pp 763–769 | Cite as

The relationship between sedative drug utilization and outcomes in critically ill patients undergoing mechanical ventilation

  • Kenshi Hayashida
  • Takeshi Umegaki
  • Hiroshi Ikai
  • Genki Murakami
  • Masaji Nishimura
  • Yuichi Imanaka
Original Article

Abstract

Objectives

The objectives of this study were to describe current sedative drug utilization patterns in critically ill patients undergoing mechanical ventilation (MV) in intensive care units (ICUs) in Japanese hospitals and to elucidate the relationship of these utilization patterns with patient clinical outcomes.

Method

Analysis of hospital claims data derived from the Quality Indicator/Improvement Project identified 12,395 critically ill adult patients who had undergone MV while hospitalized in the ICUs of 114 Japanese hospitals and had been discharged between April 2008 and March 2010. Descriptive statistics were calculated for the daily utilization of sedative drugs, opioids, and muscle relaxants in this patient sample, and the relationship between drug utilization and patient outcomes using Cox proportional hazards analysis were examined.

Results

Of the 12,395 patients included in the analysis, 7300 (58.9 %), 580 (4.7 %), and 671 (5.4 %) received sedative drugs, opioids, and muscle relaxants, respectively, for ≥2 days after intubation. Compared to the other patient groups, there was a higher proportion of males in the group given sedative drugs and the patients were significantly younger (P < 0.001). Propofol was the most frequently used sedative drug, followed by benzodiazepines, barbiturates, and dexmedetomidine. The mortality rate was lower and ventilator weaning was earlier among patients who received only propofol than among those who received only benzodiazepines. Muscle relaxants were associated with increased duration of MV.

Conclusions

This is the first study based on a large-scale analysis in Japan to elucidate sedative drug utilization patterns and their relationship with outcomes in critically ill patients. The most commonly used sedative was propofol, which was associated with favorable patient outcomes. Further prospective research must be conducted to discern effective sedative drug utilization.

Keywords

Mechanical ventilation Sedative drugs Opioids Muscle relaxants Mortality Intensive care units Japan 

Introduction

The widespread administration of sedative drugs to critically ill patients receiving mechanical ventilation (MV) has led to growing concerns among clinicians and researchers regarding their clinical advantages and disadvantages. Propofol is generally considered to be a first-line sedative [1] and is more commonly used than the benzodiazepines or dexmedetomidine [2]. Dexmedetomidine, however, has been reported to be safer and more effective than propofol [3, 4]. There is a wide range of sedative drugs available for use in the clinical setting, and an inappropriate choice may result in unfavorable clinical outcomes. The effectiveness of these drugs also has to be taken into consideration in the framework of economics due to the high medical costs associated with treating critically ill patients. The provision of necessary treatments is essential and unavoidable; however, non-essential and costly treatments place needless financial burdens on patients, hospitals, and payers. Sedative drugs contribute substantially to the medical expenditures of critically ill patients, thereby resulting in higher costs [5]. If the administration of these drugs is found to have little or no benefit on patient outcomes, they may represent a possible target for cost containment without any detrimental effect on healthcare quality.

The publication of clinical guidelines [6, 7] has led to the establishment of evidence-based medicine in many countries, including Japan. Clinical practice in Japanese intensive care units (ICUs) is shifting toward an emphasis on evidence-based medicine. However, many of these guidelines are based on the results of research conducted in countries other than those in which they are applied, and it is possible that intrinsic differences in patient characteristics and clinical patterns between countries can lead to applicability issues. Therefore, in order to support the formulation and establishment of guidelines in Japan for sedative drug use, there is a need to first generate baseline data on patient characteristics and sedative drug utilization patterns in critically ill patients receiving MV in Japanese ICUs. To the best of our knowledge, no comprehensive study has yet addressed this issue.

The aim of this study was to describe current sedative drug utilization patterns in critically ill patients receiving MV in ICUs in Japanese hospitals and to elucidate their relationship with patient clinical outcomes.

Methods

Data sources and patient selection

Hospital claims data for analysis were retrieved from the Quality Indicator/Improvement Project (QIP; http://med-econ.umin.ac.jp/QIP/), which was established in 1995 to evaluate and improve the quality of healthcare in participant hospitals in Japan. This database includes a hospital characteristic variable (hospital code), patient demographic variables (age and gender), hospitalization and release variables (admission/discharge dates and patient outcomes), diagnostic information variables (primary diagnosis, comorbidities, and complications), and clinical practice variables (procedures, intravenous drugs, and prescriptions). These data are provided in a format corresponding to the Japanese Diagnosis Procedure Combination patient classification system [8].

Critically ill patients aged 20 years and older who had been discharged between April 2008 and March 2010 from 114 QIP participant hospitals equipped with ICUs were considered to be eligible for entry into the study. The specific target group of the analysis was those patients who had received MV while in the ICU.

Target medications

The use of sedative drugs, opioids, and muscle relaxants target medications were studied. Specifically, we studied the use of propofol, barbiturates, benzodiazepine, and dexmedetomidine from the sedative drug class, fentanyl, morphine, and ketamine from the opioid drug class, and pancuronium, rocuronium, and vecuronium from the class of muscle relaxants.

Statistical analyses

As a first step, we calculated the descriptive statistics of the daily utilization of sedative drugs, opioids, and muscle relaxants for the 12,395 critically ill adult patients who met the inclusion criteria. Next, we examined the relationship between drug utilization and patient outcomes, with the latter analysis based on the reported outcomes at the time of hospital discharge and the duration of time on MV. It became evident that we were unable to distinguish between whether the administration of sedative drugs on the first day in the ICU was for sedation at the point of intubation or for MV; consequently, we identified patients who had been given sedative drugs while receiving MV as those who had been administered the target drugs for at least 2 days after MV initiation.

The statistical methods are as follows:
  1. 1.

    Patient characteristics were compared using t tests and Chi-square tests (where applicable) among the different categories of sedative drugs used on the day after intubation.

     
  2. 2.

    Descriptive statistics were calculated for the single and combined administration of sedative drugs, opioids, and muscle relaxants for the following time periods: from 1 to 7, 10, and 14 days after intubation.

     
  3. 3.

    Cox proportional hazards analysis was used to assess differences in mortality between patients who used benzodiazepines and those who used propofol, the two most frequently administered sedative drugs in our patient sample. In this analysis, five variables (age, unplanned admission, MV, hospital category, and admission diagnosis) were used as risk-adjustment covariates to account for variations in patient severity. These variables are those used in the Critical Care Outcome Prediction Equation (COPE) model [9] to adjust for patient severity.

     
  4. 4.

    Cox proportional hazards analysis was also used to assess differences in the duration of MV utilization among the four sedative drugs and the combined administration of sedative drugs and muscle relaxants. The five variables (age, unplanned admission, MV, hospital category, and admission diagnosis), which were also used in the Cox proportional hazards analysis described above, were included as risk-adjustment covariates to account for variations in patient severity.

     

All analyses were performed using the SPSS statistical package version 21.0 (SPSS/IBM Corp., Armonk, NY). Statistical significance was set at P < 0.05.

Ethical standard

This study was reviewed and approved by the Ethics committee of Kyoto University Graduate School of Medicine.

Results

A search of the original dataset resulted in the identification of 18,924 patients from among the 75,069 who had undergone MV while in the ICU. Of these patients, 12,395 had continued to receive MV on the second day after intubation, and among these 12,395 patients, 7300 (58.9 %) received sedative drugs. Table 1 shows the comparisons between patients who had been administered sedative drugs and those who had not. In general, the group which were given sedative drugs were younger and a higher proportion were male than those who did not receive the drugs (P < 0.001). Patients who received sedative drugs also had a higher proportion of primary diagnoses associated with hepatologic/immunologic, neoplastic, cardiovascular, or gastrointestinal causes as well as a higher utilization of therapies, such as continuous renal replacement therapy, endotoxin adsorption, dopamine use, noradrenaline use, and central venous catheter use.
Table 1

Characteristics of patients who had or had not received sedative drugs on the second day of mechanical ventilation

Variables

Received sedative drugs (N = 7300)

Not received sedative drugs (N = 5095)

P value

Age (years)

68.9 ± 13.8

71.2 ± 13.3

<0.001

Sex (male/female)

4745/2555

2940/2155

<0.001

Admission route (planned/emergency)a

1600/5698

1155/3940

0.33

Primary disease

  Neuromuscular

1300 (17.8 %)

999 (13.7 %)

0.01

  Respiratory

2508 (34.4 %)

1656 (22.7 %)

0.03

  Renal

1238 (17.0 %)

868 (11.9 %)

0.92

  Hematologic/Immunologic

966 (13.2 %)

552 (7.6 %)

<0.001

  Metabolic

2396 (32.8 %)

1574 (21.6 %)

0.03

  Genetic defect/Other congenital

76 (1.0 %)

61 (0.8 %)

0.43

  Neoplastic

1201 (16.5 %)

724 (9.9 %)

0.001

  Cardiovascular

5088 (69.7 %)

3749 (51.4 %)

<0.001

  Gastrointestinal

2002 (27.4 %)

1089 (14.9 %)

<0.001

  Trauma

394 (5.4 %)

341 (4.7 %)

0.003

  Toxin

142 (1.9 %)

122 (1.7 %)

0.10

Resource use

  Continuous renal replacement therapy

1250 (17.1 %)

534 (7.3 %)

<0.001

  Plasma exchange

83 (1.1 %)

44 (0.6 %)

0.14

  Endotoxin adsorption

479 (6.6 %)

194 (2.7 %)

<0.001

  Dopamine

4777 (65.4 %)

3073 (42.1 %)

<0.001

  Dobutamine

2517 (34.5 %)

1417 (19.4 %)

<0.001

  Noradrenaline

3376 (46.2 %)

2088 (28.6 %)

<0.001

  Vasopressin

248 (3.4 %)

187 (2.6 %)

0.43

  Adrenaline

1657 (22.7 %)

1216 (16.7 %)

0.13

  Enteral nutrition

3505 (48.0 %)

1561 (21.4 %)

<0.001

  Parenteral nutrition

5764 (79.0 %)

3062 (41.9 %)

<0.001

  Central venous catheter

5997 (82.2 %)

3390 (46.4 %)

<0.001

Data are presented in table as the absolute number with/without the percentage in parenthesis, with the probability calculated using the Chi-square test or as the mean ± standard deviation as determined by the t test

aMissing data in 2 cases of patients who had been given sedative drugs

Table 2 shows the utilization of sedative drugs, opioids, and muscle relaxants during post-intubation days 1–7, 10, and 14. The results showed that 7300 (58.9 %), 580 (4.7 %), and 671 (5.4 %) patients continued to receive sedative drugs, opioids, and muscle relaxants, respectively, on the second day after intubation. Regardless of how many days had passed after intubation, propofol was the most frequently used sedative drug, followed in order of decreasing usage by benzodiazepines, barbiturates, and dexmedetomidine.
Table 2

Utilization of sedative drugs, opioids, and muscle relaxants by patients receiving mechanical ventilation

Days after mechanical ventilation

1

2

3

4

5

6

7

10

14

Number of the patients undergoing mechanical ventilation

18,924

12,395

9089

7681

6658

5738

4599

4221

3198

Sedative drugs (individual use or concomitant use)

11,033 (58.3 %)

7300 (58.9 %)

5612 (61.7 %)

4641 (60.4 %)

3790 (56.9 %)

3165 (55.2 %)

2453 (53.3 %)

2059 (48.8 %)

1352 (42.3 %)

  Propofol

7897 (41.7 %)

4483 (36.2 %)

3196 (35.2 %)

2620 (34.1 %)

2113 (31.7 %)

1749 (30.5 %)

1317 (28.6 %)

1082 (25.6 %)

751 (23.5 %)

  Barbiturates

673 (3.6 %)

210 (1.7 %)

159 (1.7 %)

126 (1.6 %)

87 (1.3 %)

78 (1.4 %)

44 (1.0 % )

43 (1.0 %)

18 (0.6 %)

  Benzodiazepine

5849 (30.9 %)

3222 (26.0 %)

2619 (28.8 %)

2219 (28.9 %)

1860 (27.9 %)

1570 (27.4 %)

1263 (27.5 %)

1077 (25.5 %)

677 (21.2 %)

  Dexmedetomidine

30 (0.2 %)

27 (0.2 %)

38 (0.4 %)

28 (0.4 %)

31 (0.5 %)

32 (0.6 %)

11 (0.2 %)

15 (0.4 %)

5 (0.2 %)

Opioids (individual use or concomitant use)

1846 (9.8 %)

580 (4.7 %)

452 (5.0 %)

397 (5.2 %)

328 (4.9 %)

237 (4.1 %)

175 (3.8 %)

165 (3.9 %)

108 (3.4 %)

  Fentanyl

1398 (7.4 %)

603 (4.9 %)

497 (5.5 %)

408 (5.3 %)

344 (5.2 %)

307 (5.4 %)

147 (3.2 %)

147 (3.5 %)

103 (3.2 %)

  Morphine

607 (3.2 %)

131 (1.1 %)

100 (1.1 %)

107 (1.4 %)

71 (1.1 %)

65 (1.1 %)

23 (0.5 %)

15 (0.4 %)

14 (0.4 %)

  Ketamine

138 (0.7 %)

31 (0.3 %)

22 (0.2 %)

17 (0.2 %)

13 (0.2 %)

14 (0.2 %)

7 (0.2 %)

5 (0.1 %)

1 (0.0 %)

Muscle relaxants (individual use or concomitant use)

4212 (22.3 %)

671 (5.4 %)

444 (4.9 %)

321 (4.2 %)

271 (4.1 %)

188 (3.3 %)

141 (3.1 %)

98 (2.3 %)

81 (2.5 %)

  Pancuronium

412 (2.2 %)

64 (0.5 %)

41 (0.5 %)

37 (0.5 %)

39 (0.6 %)

25 (0.4 %)

9 (0.2 %)

10 (0.2 %)

11 (0.3 %)

  Rocuronium

1755 (9.3 %)

233 (1.9 %)

142 (1.6 %)

124 (1.6 %)

95 (1.4 %)

81 (1.4 %)

34 (0.7 %)

27 (0.6 %)

25 (0.8 %)

  Vecuronium

2462 (13.0 %)

588 (4.7 %)

391 (4.3 %)

285 (3.7 %) 

242 (3.6 %)

206 (3.6 %)

98 (2.1 %)

61 (1.4 %)

58 (1.8 %)

Data are presented in table as the absolute number of patients with the percentage in parenthesis

Table 3 shows the concomitant administration of opioids and muscle relaxants to patients who had been given a sedative drug during the post-intubation period on days 1–7, 10, and 14. The results show that fentanyl (opioid) and vecuronium (muscle relaxant) were administered in combination with all sedative drugs. Approximately 5–8 % of patients with a single administration of propofol and approximately 9–12 % of patients with benzodiazepine received some opioids. In comparison, approximately 3–6 % of patients with a single administration of propofol and approximately 5–13 %  of patients with benzodiazepine received some muscle relaxants.
Table 3

Concomitant use of opioids and muscle relaxants for patients who received sedative drugs

Sedative drugs

Number of days after mechanical ventilation

1

2

3

4

5

6

7

10

14

Propofol

4263

3290

2455

1998

1586

1310

973

828

588

  Concomitant use with:

    Fentanyl

309 (7.2 %)

168 (5.1 %)

129 (5.3 %)

106 (5.3 %)

98 (6.2 %)

64 (4.9 %)

49 (5.0 %)

57 (6.9 %)

43 (7.3 %)

    Morphine

100 (2.3 %)

15 (0.5 %)

14 (0.6 %)

10 (0.5 %)

10 (0.6 %)

7 (0.5 %)

5 (0.5 %)

2 (0.2 %)

3 (0.5 %)

    Ketamine

11 (0.3 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

2 (0.2 %)

0 (0.0 %)

0 (0.0 %)

    Pancuronium

13 (0.3 %)

7 (0.2 %)

0 (0.0 %)

5 (0.3 %)

0 (0.0 %)

2 (0.2 %)

0 (0.0 %)

2 (0.2 %)

0 (0.0 %)

    Rocuronium

318 (7.5 %)

30 (0.9 %)

18 (0.7 %)

14 (0.7 %)

12 (0.8 %)

9 (0.7 %)

13 (1.3 %)

10 (1.2 %)

14 (2.4 %)

    Vecuronium

435 (10.2 %)

112 (3.4 %)

76 (3.1 %)

55 (2.8 %)

43 (2.7 %)

31 (2.4 %)

25 (2.6 %)

10 (1.2 %)

20 (3.4 %)

Barbiturates

153

109

90

76

52

50

26

20

8

  Concomitant use with:

    Fentanyl

12 (7.8 %)

2 (1.8 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Morphine

5 (3.3 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Ketamine

1 (0.7 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Pancuronium

1 (0.7 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Rocuronium

22 (14.4 %)

1 (0.9 %)

0 (0.0 %)

0 (0.0 %)

2 (3.8 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Vecuronium

20 (13.1 %)

10 (9.2 %)

5 (5.6 %)

3 (3.9 %)

4 (7.7 %)

4 (8.0 %)

2 (7.7 %)

0 (0.0 %)

1 (12.5 %)

Benzodiazepine

2416

2029

1704

1418

1156

995

769

683

420

  Concomitant use with:

    Fentanyl

163 (6.7 %)

156 (7.7 %)

131 (7.7 %)

116 (8.2 %)

90 (7.8 %)

82 (8.2 %)

51 (6.6 %)

54 (7.9 %)

41 (9.8 %)

    Morphine

67 (2.8 %)

31 (1.5 %)

17 (1.0 %)

17 (1.2 %)

16 (1.4 %)

14 (1.4 %)

10 (1.3 %)

12 (1.8 %)

6 (1.4 %)

    Ketamine

15 (0.6 %)

12 (0.6 %)

10 (0.6 %)

6 (0.4 %)

0 (0.0 %)

2 (0.2 %)

4 (0.5 %)

2 (0.3 %)

3 (0.7 %)

    Pancuronium

32 (1.3 %)

19 (0.9 %)

13 (0.8 %)

11 (0.8 %)

11 (1.0 %)

7 (0.7 %)

5 (0.7 %)

2 (0.3 %)

5 (1.2 %)

    Rocuronium

192 (7.9 %)

60 (3.0 %)

46 (2.7 %)

36 (2.5 %)

25 (2.2 %)

13 (1.3 %)

12 (1.6 %)

9 (1.3 %)

8 (1.9 %)

    Vecuronium

361 (14.9 %)

184 (9.1 %)

120 (7.0 %)

89 (6.3 %)

73 (6.3 %)

47 (4.7 %)

34 (4.4 %)

26 (3.8 %)

26 (6.2 %)

Dexmedetomidine

9

9

11

11

15

14

1

8

2

  Concomitant use with:

    Fentanyl

0 (0.0 %)

0 (0.0 %)

1 (9.1 %)

0 (0.0 %)

2 (13.3 %)

3 (21.4 %)

0 (0.0 %)

1 (12.5 %)

0 (0.0 %)

    Morphine

1 (11.1 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Ketamine

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Pancuronium

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Rocuronium

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

    Vecuronium

0 (0.0 %)

1 (11.1 %)

1 (9.1 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

0 (0.0 %)

Data are presented as the absolute number of patients with the percentage in parenthesis

Table 4 shows the relationship between the administration of common sedative drugs and mortality. A higher mortality rate was observed among patients who received only benzodiazepine compared to those who received only propofol (odds ratio 1.23; 95 % confidence interval 1.13–1.35).
Table 4

Relationship between different sedative drugs and mortality

Independent variables

Hazard ratio (95 % Confidence interval)

P value

Severity

2.26 (1.81–2.83)

<0.001

Sedative drug usage

  Propofol

1.00

 

  Benzodiazepine

1.23 (1.13–1.35)

<0.001

Table 5 shows the relationship between the administration of common sedative drugs and MV duration. Ventilator weaning occurred earlier in patients who received only propofol than in those who had received any other sedative drug. The administration of muscle relaxants was associated with increased duration of MV.
Table 5

Relationship between different sedative drugs and mechanical ventilation duration (until extubation)

Independent variables

Hazard ratio (95 % Confidence interval)

P value

Severity

0.55 (0.48–0.63)

<0.001

Sedative drug usage

  Propofol (individual use)

1.00

 

  Barbiturates (individual use)

0.81 (0.67–0.98)

0.033

  Benzodiazepine (individual use)

0.83 (0.79–0.87)

<0.001

  Dexmedetomidine (individual use)

1.25 (0.65–2.40)

0.512

  Propofol and benzodiazepine

0.91 (0.83–1.01)

0.064

  Propofol and barbiturates

0.61 (0.39–0.96)

0.032

  Propofol and dexmedetomidine

0.89 (0.53–1.51)

0.671

  Benzodiazepine and barbiturates

0.82 (0.62–1.09)

0.165

Concomitant use with other drugs

  Sedative drugs and muscle relaxants

0.82 (0.75–0.90)

<0.001

Discussion

To the best of our knowledge, this is the first large-scale analysis of Japanese hospitals to elucidate sedative drug utilization patterns and their relationship with outcomes in critically ill patients receiving MV in ICUs. Our results show that the administration of propofol was associated with earlier extubation and decreased mortality and that the administration of muscle relaxants was associated with increased duration of MV. These results indicate a need to further investigate the proper utilization of muscle relaxants, specifically focusing on their possible elimination as therapy for critically ill patients undergoing MV.

The observed association between propofol use and lower mortality or earlier extubation corroborates the results of previous studies. Hannah et al. reported that patients receiving only propofol had lower mortality rates than those who received only benzodiazepines [2]. In a Canadian multicenter study, Hall et al. compared MV duration between patients who had been administered propofol and those receiving midazolam and concluded that the administration of propofol sedation resulted in more rapid tracheal extubation than midazolam sedation [10]. In a review study, Fulton and Sorkin concluded that propofol-sedated patients tended to have a faster time to spontaneous ventilation or extubation than midazolam-sedated patients [11]. Anger et al. evaluated dexmedetomidine and propofol use in mechanically ventilated cardiac surgery patients and reported no difference in MV duration between the two patient groups [12]. However, comparisons of the relative superiority of dexmedetomidine and propofol remain a matter of contention due to the lack of cases involving dexmedetomidine sedation. In view of its relatively recent introduction onto the market, dexmedetomidine utilization should be analyzed carefully in the near future. Although initial utilization rates were low, this sedative drug is now being used with increasing frequency.

Studies have revealed that in terms of the economic burden of administering sedative drugs, there are no clear advantages between propofol or midazolam. In their review study, McKeage and Perry reported that the propofol administration was associated with a lower financial burden than midazolam administration [13]. Anis et al. reported that because propofol-sedated patients tend to be extubated earlier than those sedated with midazolam, total ICU costs for the former are correspondingly lower [14]. However, these authors also found that despite the earlier extubation of propofol-sedated patients, there were no differences between the two patient groups with respect to total hospitalization costs per patient, which include drug costs and ICU costs [14]. Further research is clearly required to clarify this issue.

Some limitations must be considered when interpreting the results of our study. First, we have analyzed the relationships between drug usage and outcomes, but the results do not provide insight into the causal direction of the relationships identified. Second, there is currently no universally accepted standard for determining the use or non-administration of sedative drug usage. Therefore, the appropriateness of our choice to use drug utilization on the second day after intubation to identify patients receiving sedative drugs while on MV may not be optimal; as well, the use of this definition to examine the relationship between the types of common sedative drugs and patient outcomes is arguable. However, we postulate that our method of identifying sedative drug use while the patient is receiving continuous MV is appropriate in the context of the data available. As it is common practice to use sedative drugs when performing intubation at the initiation of MV, we decided not to focus on sedative drug use on the first day of MV for our analysis. Also, we believe that it is acceptable to assume that the first choice drug for sedation is representative of the general sedation policy for patients receiving MV. Third, more recent sedative drug utilization patterns may differ from those during our study period. In Japan, dexmedetomidine became covered by health insurance in 2004 and was initially used only in patients who had a high chance of early extubation. In 2010, however, its application was expanded to all patients receiving MV in the ICU. Thus, dexmedetomidine utilization may be more widespread at the present time than during our study period, but this difference is unlikely to have resulted in dexmedetomidine utilization surpassing that of propofol. Fourth, due to the nature of our database, some patients were sedated for 2 days using the first day’s prescription of sedative drugs, which would appear as a lack of sedative drug utilization on the second day in our data. Although this confounding factor may have led to an underestimation of the use of sedative drugs in our study, our overall conclusions are unlikely to be affected even if actual drug utilization was higher. For example, the general utilization tendencies across all sedative drugs were likely to be similar even if there were slight variations in individual drug utilization rates. Also, by focusing only on patients who used sedative drugs, this limitation would have minimal influence on the observed associations between sedative drug utilization and outcomes. Future analyses should incorporate clinical data to validate these findings. Finally, the hospitals selected in this study may not be representative of all hospitals in Japan. In particular, our sample did not include any major university hospitals.

Conclusions

We were able to successfully shed light on the patient characteristics, clinical patterns in sedative drug utilization, and the relationship between the types of common sedative drugs and patient outcomes in critically ill patients receiving MV in Japanese ICUs. We were unable to determine a causal connection. Consequently, there is a need for further prospective research to be conducted to identify effective sedative drug utilization policies for critically ill patients undergoing MV and to investigate the formulation and revision of clinical guidelines for sedative drugs that are best suited for Japanese patients.

Notes

Acknowledgments

This work was financially supported in part by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science ([A]25253033).

Compliance with ethical standards

Conflict of interest

The authors have no conflicts of interests to declare.

References

  1. 1.
    McGrane S, Pandharipande PP. Sedation in the intensive care unit. Minerva Anestesiol. 2012;78:369–80.PubMedGoogle Scholar
  2. 2.
    Wunsch H, Kahn JM, Kramer AA, Rubenfeld GD. Use of intravenous infusion sedation among mechanically ventilated patients in the United States. Crit Care Med. 2009;37:3031–9.CrossRefPubMedGoogle Scholar
  3. 3.
    Venn RM, Grounds RM. Comparison between dexmedetomidine and propofol for sedation in the intensive care unit: patient and clinician perceptions. Br J Anaesth. 2001;87:684–90.CrossRefPubMedGoogle Scholar
  4. 4.
    Herr DL, Sum-Ping ST, England M. ICU sedation after coronary artery bypass graft surgery: dexmedetomidine-based versus propofol-based sedation regimens. J Cardiothorac Vasc Anesth. 2003;17:576–84.CrossRefPubMedGoogle Scholar
  5. 5.
    Kress JP, Pohlman AS, O’Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med. 2000;342:1471–7.CrossRefPubMedGoogle Scholar
  6. 6.
    Jacobi J, Fraser GL, Coursin DB, Riker RR, Fontaine D, Wittbrodt ET, Chalfin DB, Masica MF, Bjerke HS, Coplin WM, Crippen DW, Fuchs BD, Kelleher RM, Marik PE, Nasraway SA Jr, Murray MJ, Peruzzi WT, Lumb PD, Task Force of the American College of Critical Care Medicine (ACCM) of the Society of Critical Care Medicine (SCCM), American Society of Health-System Pharmacists (ASHP), American College of Chest Physicians. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med. 2002;30:119–41.CrossRefPubMedGoogle Scholar
  7. 7.
    Barr J, Fraser GL, Puntillo K, Ely EW, Gelinas C, Dasta JF, Davidson JE, Devlin JW, Kress JP, Joffe AM, Coursin DB, Herr DL, Tung A, Robinson BR, Fontaine DK, Ramsay MA, Riker RR, Sessler CN, Pun B, Skrobik Y, Jaeschke R, American College of Critical Care Medicine. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013;41:263–306.CrossRefPubMedGoogle Scholar
  8. 8.
    Matsuda S, Ishikawa KB, Kuwabara K, Fujimori K, Fushimi K, Hashimoto H. Development and use of the Japanese case-mix system. Eurohealth. 2008;14:25–30.Google Scholar
  9. 9.
    Duke GJ, Santamaria J, Shann F, Stow P, Pilcher D, Ernest D, George C. Critical care outcome prediction equation (COPE) for adult intensive care. Crit Care Resusc. 2008;10:41.PubMedGoogle Scholar
  10. 10.
    Hall RI, Sandham D, Cardinal P, Tweeddale M, Moher D, Wang X. Anis AH; Study Investigators. Propofol vs midazolam for ICU sedation: a Canadian multicenter randomized trial. Chest. 2001;119:1151–9.CrossRefPubMedGoogle Scholar
  11. 11.
    Fulton B, Sorkin EM. Propofol. An overview of its pharmacology and a review of its clinical efficacy in intensive care sedation. Drugs. 1995;50:636–57.CrossRefPubMedGoogle Scholar
  12. 12.
    Anger KE, Szumita PM, Baroletti SA, Labreche MJ, Fanikos J. Evaluation of dexmedetomidine versus propofol-based sedation therapy in mechanically ventilated cardiac surgery patients at a tertiary academic medical center. Crit Pathw Cardiol. 2010;9:221–6.CrossRefPubMedGoogle Scholar
  13. 13.
    McKeage K, Perry CM. Propofol: a review of its use in intensive care sedation of adults. CNS Drugs. 2003;17:235–72.CrossRefPubMedGoogle Scholar
  14. 14.
    Anis AH, Wang XH, Leon H, Hall R, Propofol Study Group. Economic evaluation of propofol for sedation of patients admitted to intensive care units. Anesthesiology. 2002;96:196–201.CrossRefPubMedGoogle Scholar

Copyright information

© Japanese Society of Anesthesiologists 2016

Authors and Affiliations

  • Kenshi Hayashida
    • 1
  • Takeshi Umegaki
    • 2
  • Hiroshi Ikai
    • 3
  • Genki Murakami
    • 1
  • Masaji Nishimura
    • 4
  • Yuichi Imanaka
    • 3
  1. 1.Department of Medical Informatics and Management, University HospitalUniversity of Occupational and Environmental HealthKitakyushu, FukuokaJapan
  2. 2.Department of AnesthesiologyKansai Medical UniversityHirakata, OsakaJapan
  3. 3.Department of Healthcare Economics and Quality ManagementKyoto University Graduate School of MedicineSakyo-ku, KyotoJapan
  4. 4.Department of Emergency and Critical Care MedicineTokushima University Graduate SchoolTokushimaJapan

Personalised recommendations