Advice

Appendix

Appendix

Contents

Data tables

Table 1: Overview of the Holzinger et al. (2011) trial

Table 2: Summary of results from the Holzinger et al. (2011) trial

Table 3: Overview of the Gray et al. (2007) study

Table 4: Overview of the October and Hardart (2009) study

Table 5: Overview of the Kline et al. (2011) trial

Table 6: Overview of the Viana et al. (2011) study

Table 7: Overview of the Powers et al. (2011) study

Table 8: Overview of the Taylor et al. (2014) study

Table 9: Summary of the economic studies (Brown et al. 2012; Gray et al. 2007; October and Hardart 2009; Sharma and Nair 2013; Taylor et al. 2010; Windle et al. 2010)

Table 1 Overview of the Holzinger et al. (2011) trial

Study component

Description

Objectives/hypotheses

To compare the success rate of NJ feeding tube placement using CORTRAK 1 EAS with that of the endoscopic technique in critically ill patients.

Study design

Single centre randomised controlled trial

Setting

Two ICUs at a university hospital in Austria; May 2007- February 2009

No follow up period was reported.

Inclusion/exclusion criteria

Inclusion:

  • >18 years of age

  • intubated, mechanically ventilated,

  • intolerance of intragastric enteral nutrition.

Exclusion:

  • contraindication for enteral nutrition or upper gastrointestinal endoscopy

  • previous upper gastrointestinal surgery

  • signs of active upper gastrointestinal bleeding

  • severe nasopharyngeal injuries or stenosis.

Primary outcomes

Primary outcome:

  • The success rate of correct NJ tube placement after 24 hours

Secondary outcomes:

  • Success rate

  • Placement time (mins)

  • Nose bleeding events

  • Number of tube placement

  • Time in right location (days)

  • Intensive care unit survival

  • Hospital survival

Statistical methods

The study aimed to detect a difference in success probability of 30% given a baseline probability of nearly 100% at a power of 80% and a significance level of 5%.

For the comparison of the "number of attempts" a chi-square test was used.

Multiple linear regression was used to adjust the continuous variables for reflux.

Patients included

In total 66 critically ill adults who were unable to tolerate intragastric nutrition were recruited.

CORTRAK 1 EAS (study group):

n=44, age=55±18 years; 64% male, 35% female.

Admission reason: 9% septic shock, 18% heart failure, 21% acute lung injury/adult respiratory distress syndrome, 41% burn injury, 11% others.

Endoscopic technique using a 150 cm double lumen jejunal tube (Freka Trelumina, manufactured by Fresenius Kabi; control group):

n=22, age=56±15 years; 82% male, 18% female.

Admission reason: 9% septic shock, 23% heart failure, 27% acute lung injury/adult respiratory distress syndrome, 36% burn injury, 5% others.

Results

Correct NJ tube placement was achieved in 21 of 22 patients using the endoscopic technique (control group) and in 40 of 44 patients using CORTRAK (intervention group) (p=0.571).

The placement times were not significantly different between the two groups (p=0.23).

CORTRAK resulted in the correct nasojejunal position more often at the first attempt (p=0.009).

Nose bleeding occurred equally in both groups (p=0.99). None of the nose bleeding events needed intervention or transfusion.

There was no significant difference between groups in the amount of time the tubes stayed in the correct location (mean difference -1.32 days, CI -2.09 to +1.21).

Neither ICU survival (p=0.72) nor hospital survival p=0.84) were different between the groups.

Conclusions

Correct NJ feeding tube placement using the CORTRAK system was as fast, safe, and successful as the endoscopic method in a comparative ICU patient population.

Abbreviations: CI, confidence interval; NJ, nasojejunal; n, number of patients; RR, relative risk.

Table 2 Summary of results from the Holzinger et al. (2011) trial

CORTRAK 1 EAS

Endoscopic technique

Analysis

Randomised

n=44

n=22

Efficacy

n=44

n=22

Primary outcome: successful implantation of the tube

91.0% (40/44)

95.0% (21/22)

RR: 0.95, 95% CI 0.80 to 1.12

p=0.57

Selected secondary outcomes:

Placement time mean and range (minutes)

11 (6-19)

15 (10-21)

mean difference -1.26 minutes, 95% CI -1.88 to +1.17, p=0.23

Number of attempts

1.18±0.54

1.82±0.79

1.18±0.54, p<0.0001

Nose bleeding

18.0% (8/44)

18.0% (4/22)

RR 1.00, 95% CI 0.01 to 140.69, p=0.99

Time left in right location mean and range (days)

9 (4.5-16.5)

13 (9-20)

p=0.27

ICU survival

72.0%(32/44)

68.0%(15/22)

RR 1.07, 95% CI 0.75 to 1.51, p=0.72

Hospital survival

61.0%(27/44)

59.0%(13/22)

RR 1.04, 95%CI 0.73 to 1.48, p=0.84 

Safety

n=44

n=22

Patients reporting serious adverse events

0% (0/44)

0.0% (0/22)

Abbreviations: CI, confidence interval; ICU, intensive care unit; n, number of patients; RR, relative risk.

Table 3 Overview of the Gray et al. (2007) study

Study component

Description

Objectives/hypotheses

To evaluate the safety of CORTRAK 1 EAS for placing feeding tubes compared with blind placement.

Study design

Single centre prospective observational study

Setting

ICU (USA)

Control group recruitment: February 2004 to May 2005 

Study group recruitment: August 2005 to March 2006

No follow up period was reported.

Inclusion/exclusion criteria

Inclusion:

  • ICU patients who needed post-pyloric feeding tubes

Exclusion:

  • implantable devices: automatic internal cardiac defibrillator (AICD), cardiac or gastric pacemaker

  • head trauma involving sinuses or nares

  • severe bleeding/clotting disorders such as disseminated intravascular coagulation (DIC).

Primary outcomes

Primary outcome:

  • safety/complications related to post-pyloric tube placement

Secondary outcomes:

  • success rate of accurate placement into the small intestine

  • number of X-rays needed to confirm placement

  • basic cost of post-pyloric tube placement

  • time from clinician request to placement

  • time from placement to initiation of enteral feeding.

Statistical methods

Statistical analysis was performed using the Wilcoxon signed rank test to compare:

  • number of abdominal X-rays needed to confirm placement.

  • basic cost of post-pyloric tube placement

  • time from clinician request to placement

  • time from placement to feeding

The post-pyloric tube placement success rate was analysed using chi-squared and Fisher's exact test.

Patients included

A total of 101 adults were recruited from ICU.

CORTRAK 1 EAS (intervention group):

n=81 (mean age: 55.9 years, 31% female, 69% male).

Admission reason: 5% cardiovascular surgery, 10% gastrointestinal, 12% haematology/oncology, 14% sepsis/organ failure, 18% neurology/trauma, 18% organ transplant, 23% respiratory failure.

Blind placement (control group): n=20 (mean age: 53.8 years, 60% female, 40% male).

Admission reason: 5% neurology/trauma, 10% haematology/oncology, 20% organ transplant, 25% respiratory failure, 40% sepsis/organ failure.

Results

Primary outcome:

No complications or adverse events were reported in the intervention or the control groups.

Secondary outcomes:

• Successful placement of the post-pyloric tubes into the small intestine was 78% in the CORTRAK group vs 63% in the blind placement group (not significant).

• Median number of X-rays was 1 in the CORTRAK group vs 2 in the control group (p=0.0001).

• The total time between clinician request for post-pyloric tube placement and initiation of feeding was 66% lower in the CORTRAK group versus the control group (median time of 7.75 hours versus 22.25 hours, p=0.0032).

• Median time between tube placement and initiation of enteral feeding was 4.0 hours in the CORTRAK group and 4.5 hours in the control group (not significant).

• The time between clinician request (for tube placement) and actual tube placement was 2.5 hours for CORTRAK) versus 4.75 hours (controls) (p=0.0059).

Conclusions

No adverse events occurred during post-pyloric tube placement in the intervention or control group. The use of CORTRAK resulted in fewer X-rays for confirmation of placement and more timely initiation of enteral feeding compared with blind placement.

Abbreviations: DIC, disseminated intravascular coagulation; GERD, gastroesophageal reflux disease; INR, international normalized ratio; ICU, intensive care unit; n, number of patients; PT, prothrombin time; PTT, partial thromboplastin time; USA, United States of America.

Table 4 Overview of the October and Hardart (2009) study

Study component

Description

Objectives/hypotheses

To evaluate the clinical and cost effectiveness of CORTRAK 1 EAS for bedside post-pyloric tube placement in comparison with standard blind placement.

Study design

Prospective cohort trial with serial control groups

Setting

Paediatric ICU at a tertiary care children's hospital (USA); from September 2005 to April 2006

No follow up period was reported.

Inclusion/exclusion criteria

Inclusion:

  • children in ICU who underwent post-pyloric tube placement between 7am and 7pm.

Exclusion:

  • weight <5 kilograms

  • post-pyloric tube placement between 7pm and 7am

  • presence of an internal cardiac pacemaker or defibrillator device, or external cardiac pacing during post-pyloric tube placement.

Primary outcomes

Primary outcome: success rate of tube placement as determined by abdominal X-ray.

  • success overall

Secondary outcomes:

  • correct location of the tube

  • number of abdominal X-rays needed to confirm tube placement,

  • time to success placement (hours)

  • complications,

  • number of tubes used,

  • prokinetic drug use.

Statistical methods

Categorical variables were assessed by chi-square or Fisher's exact test.

Continuous variables were analysed using unpaired Student's t test for parametric data and the Wilcoxon signed rank test for nonparametric data.

Ordinal variables were analysed with the Cochran-Armitage Test.

Two logistic regression models were performed; 1 with a response variable of successful outcome and 1 with successful placement dichotomized to <24 hours or >24 hours.

Multivariate regression analyses were used to adjust for more than one variable. Cumulative-event curves were estimated using the Kaplan-Meier method and the treatment groups were compared using the log-rank test.

All statistical tests were two-sided with a significance level of 0.05.

Patients included

A total of 107 children in ICU were recruited.

CORTRAK 1 EAS (intervention group): n=50, age (months): median (range), 31 (1-211); 48% male, 52% female.

Admission reason: 24% cardiac (medical and surgical); 16% neuromuscular disease; 50% respiratory failure; 10% sepsis; 0% trauma.

Blind placement (control group): n=57; age (months): median (range), 39 (2-296); 61% male, 39% female.

Admission reason: 38.6% cardiac (medical and surgical); 12.3% neuromuscular disease; 35.1% respiratory failure; 8.8% sepsis; 5.3% trauma.

Results

CORTRAK 1 EAS vs blind placement technique:

Success rate (% patients): 82.0% vs 39.0% (p<0.0001)

Placement success in <24 hours (% patients): 78.0% versus 21.1% (p<0.0001)

Median time to successful placement including confirmation (range): 1.7 hours (0.2-130 hours) versus 21.0 hours (1-477 hours) (p<0.0001).

Significantly fewer abdominal X-rays were needed in the CORTRAK 1 EAS group (mean, SE) (1.3±0.6) than in the controls (mean, SE) (2.4±1.4) (p<0.0001).

There was no difference in location (duodenum) of successful post-pyloric tube placements between the 2 groups.

CORTRAK 1 EAS was the only predictor of successful placement (odds ratio 8.4; 95% CI 2.6 –27.8) and early successful placement (success within 24 hours) (odds ratio 13.3; 95% CI 5.3–33.5) by univariate analysis.

No difference was detected in the time to initiation of feeding between the 2 groups.

There were no acute complications during the trial period in either group.

Conclusions

CORTRAK 1 EAS is an effective system for bedside post-pyloric tube placement in critically ill children.

Abbreviations: CI, confidence interval; n, number of patients; ICU, intensive care unit; SE, standard error.

Table 5 Overview of the Kline et al. (2011) trial

Study component

Description

Objectives/hypotheses

To determine whether CORTRAK 1 EAS shortens the time needed to achieve accurate placement of post-pyloric feeding tubes in critically ill children when compared with the standard blind placement technique.

Study design

Single centre randomised controlled trial

Setting

Paediatric ICU (USA)

No follow-up period was reported

Inclusion/exclusion criteria

Inclusion:

  • neonates to less than 18 years of age

  • request for placement of a post-pyloric feeding tube.

Exclusion:

  • presence of a permanent or temporary pacemaker

  • presence of a gastrostomy or jejunostomy tube

  • gastrointestinal surgery in the preceding 90 days

  • recent gastrointestinal haemorrhage

  • cardiac surgery in the preceding 2 weeks.

Primary outcomes

Primary outcome:

  • Time needed for accurate placement of a post-pyloric feeding tube.

Secondary outcome:

  • The placement success rate, as confirmed by X-rays.

Statistical methods

A Kaplan-Meier curve was used to examine differences in time to placement between the 2 groups and to estimate median time to placement, with Cox proportional hazards modelling used to determine significance.

Differences in the rate of successful placement between groups were tested by using Fisher's exact test.

Associations of time to placement with confounding factors were examined by using Cox proportional hazards modelling, again adjusting for clustering by practitioner by using robust standard errors.

Differences between the 2 methods were compared by using a chi-squared test for nominal variables and a t-test for continuous variables.

The level of significance used was 0.05.

Patients included

In total 49 children were recruited.

CORTRAK 1 EAS (intervention group):

n=22; age (mean±SD): 1.9±3.0 years; 32% male, 68% female.

Blind placement (control group):

n=27; age (mean±SD): 2.1±4.2 years; 63% male, 37% female.

Results

Primary outcome:

  • Time to placement was significantly longer in the

CORTRAK group than in the control group ((9.5 vs 5.0 minutes; HR=2.10, 95% CI 1.10 to 4.10; p=0.03).

Secondary outcome(s):

  • The success rate, as confirmed by X-rays, was 92% in the control group and 100% in the intervention group (p=0.49).

  • No adverse events were noted during tube placement in either the intervention or control group.

Conclusions

The use of CORTRAK 1 EAS does not benefit experienced practitioners in correctly placing post-pyloric feeding tubes in critically ill children when compared to the use of other routinely used techniques such as the blind placement technique.

Abbreviations: CI, confidence interval; n, number of patients; ICU, intensive care unit; HR, hazard ratio; SD, standard deviation.

Table 6 Overview of the Viana et al. (2011) study

Study component

Description

Objectives/hypotheses

The aim was to evaluate the success rate of post-pyloric feeding tube placement using CORTRAK 1 EAS compared to the conventional placement method (blind placement).

Study design

Single centre randomised controlled trial

Setting

General ICU of a tertiary hospital (Brazil)

September 2008-December 2008

No follow up period was reported.

Inclusion/exclusion criteria

Inclusion:

  • >18 years of age and at least one of the following

criteria:

  • requirement of a nasoenteral tube for enteral diet administration

  • indication for post-pyloric positioning

  • evidence of delayed gastric emptying

  • aspiration of large amounts of gastric content

  • repeated gastric aspirations

  • history of pulmonary aspiration of gastric contents

  • high-risk status

  • severe acute pulmonary disease

Exclusion:

  • active gastrointestinal bleeding

  • history of oesophageal or gastric varices

  • severe thrombocytopenia (<50,000)

  • recent oesophageal or stomach surgery

  • pharyngeal or laryngeal obstruction

  • psychomotor agitation

  • contraindications for >30º angle of the head of the bed, head or face trauma

  • requirement for non-invasive mechanic ventilation.

Primary outcomes

Primary outcome:

  • the success rate of post-pyloric placement (as verified by pH aspirations and X-ray)

Secondary outcomes:

  • time required to perform the entire procedure

Statistical methods

For continuous, normally distributed variables the Student's t-test was used. The Mann-Whitney U test was used for non-normally distributed continuous variables.

For nominal variables, the Fisher exact test was used, with confidence intervals based on a normal approximation of a binomial distribution.

Values of p<0.05 were considered significant.

Patients included

In total 37 adults were recruited.

CORTRAK 1 EAS (study group):

n=18, age=65.8±11.3 years; 44.4% male, 55.6% female.

Admission reason: clinical patients 38.9%, elective surgery 38.9%, emergency surgery 22.2%, invasive mechanical ventilation 92.3%, use of vasopressors 29.4%, use of gastric protectors 50.0%, fluid form digestive tube 47.2% aspiration, tube replacement required 5.4%.

Blind placement (control group):

n=19, age=68.7±16.7 years; 47.4% male, 52.6% female.

Admission reason: clinical patients 52.6%, elective surgery 21.1%, emergency surgery 26.3%, invasive mechanical ventilation 88.2%, use of vasopressors 47.4%, use of gastric protectors 55.6%, fluid from digestive aspiration 36.8%, tube replacement required 5.3%.

Results

Primary outcome:

  • The CORTRAK 1 EAS group achieved more post-pyloric placement than the control group (p<0.001)

  • Secondary outcomes:

  • Time required to perform the entire procedure was shorter in the CORTRAK1 EAS than in the control group (p<0.001).

Conclusions

The CORTRAK method was quicker and provided better placement accuracy than the blind placement technique.

Abbreviations: EAS, enteral access system; ICU, intensive care unit; n, number of patients.

Table 7 Overview of the Powers et al. (2011) study

Study component

Description

Objectives/hypotheses

To compare the accuracy of CORTRAK 1 EAS versus abdominal X-ray for confirmation of bedside placement of post-pyloric feeding tubes.

Study design

Multicentre prospective cohort study

Setting

The study was carried out in 3 tertiary referral centres in the USA.

No follow up period was reported.

Inclusion/exclusion criteria

Inclusion:

  • any inpatient with a request for post-pyloric feeding tube placement

Exclusion:

  • subjects who were unable to have bedside placement because of gastric bypass surgery, hiatal hernia, or bleeding complications.

  • pregnancy

  • hyperthyroidism

  • contraindications to the use of barium.

Primary outcomes

Primary outcome:

  • Accuracy of post-pyloric tube placement (% of agreement between CORTRAK 1 EAS interpretation and abdominal X-ray).

Secondary outcomes:

  • Success rate of post-pyloric feeding tube placement.

  • Percent where inadvertent airway placement was avoided.

  • Median time for feeding tube placement using CORTRAK 1 EAS.

Statistical methods

Accurate placement using the CORTRAK 1 EAS and abdominal X-rays were compared using percentage of agreement.

All variables were assessed to determine any possible correlations between gender, diagnosis, duration of placement, and investigator as predictors of proper placement or complications.

Patients included

A total of 194 people were recruited (including 18 children).

Age: Mean for all patients (range): 55 years (12 days-102 years)

60% male, 40% female

Admission reason: 50.2% medical, 25.4% neurological, 13.2% trauma, 11.2% surgery,

Admitting service: 78.4% ICU, 12.4% non-ICU, 9.2% paediatric.

Results

Primary outcome:

  • Percentage agreement between CORTRAK 1 EAS interpretation and:

    • 1st abdominal X-ray: 86.9

    • 2nd abdominal X-ray: 97.4 

    • Independent X-ray: 99.5

Secondary outcomes:

  • Successful placement at the bedside: 191 post-pyloric feeding tubes (98.4%)

  • Avoidance of inadvertent airway placement (based on detection with CORTRAK 1 EAS): 7.5% of patients,

  • Median time for feeding tube placement using CORTRAK 1 EAS in minutes (range): 12 (1–52) (1 outlier of 122 minutes).

  • No complications associated with the use of CORTRAK 1 EAS were identified.

Paediatric subset data results:

  • Median time for post-pyloric feeding tube placement (min, range): 20 (1–45).

  • Correlation between CORTRAK 1 EAS interpretation and the first X-ray: 99.4% and 100% (before and after injection of water-soluble contrast respectively)

  • Avoidance of inadvertent airway placement: 22%.

  • No complications were encountered during placement.

Conclusions

Findings indicate that the use of CORTRAK 1 EAS can accurately confirm placement of post-pyloric tubes at the bedside when compared to X-ray. CORTRAK 1 EAS can be used safely at the bedside to facilitate placement of feeding tubes.

Abbreviations: ICU, intensive care unit.

Table 8 Overview of the Taylor et al. (2014) study

Study component

Description

Objectives/hypotheses

To determine accuracy of the CORTRAK EAS in guiding NG feeding tube location when compared with pH and, in cases where results were inconclusive, X-rays.

Study design

Prospective, single centre study

Setting

ICU (UK)

No follow up period was reported.

Inclusion/exclusion criteria

Inclusion:

  • ICU adults requiring a new or replacement nasogastric tube.

Exclusion:

  • No exclusion criteria were provided

Primary outcomes

Primary outcome:

  • Success rate of nasogastric tube placement as determined by pH and X-ray

Secondary outcome:

  • Time to confirmation of placement

Statistical methods

Agreement between CORTRAK EAS NG tube placement with the other methods was tested using Cohen's kappa coefficient.

Patients included

A total of 113 adults were recruited from ICU (with a total of 127 tube placements).

Age: median (IQR); 53 (36, 66)

Weight: median (IQR); 80 kg (68, 90)

Height: median (IQR) 174 cm (166, 180)

66% male, 34% female

Admission reason: 30% medical, 12% neurosurgery, 14% surgery (general), 44% trauma.

Results

Primary outcome:

Success rate: of 127 placements, 125 of 127 placements (98%) were confirmed as correctly NG.

Secondary outcomes:

CORTRAK EAS placement and confirmation of placement took a median of 6.4 minutes (IQR: 4, 10.4) and was completed in the late morning (median 11:30: 11:00, 12:24) during the 8:00–12:00 no feeding 'rest' period, whereas X-ray, when needed, was completed later (median 14:00: 13:00, 15:00).

Confirmation of position was immediate for CORTRAK and pH but X-ray delayed feeding and medicines by 2 hours.

Conclusions

The use of CORTRAK EAS results in quick and successful placement of most NG tubes.

Abbreviations: CI, confidence interval; ICU, intensive care unit; IQR, interquartile range; NG, nasogastric; n, number of patients.

Table 9 Summary of the economic studies

Author (year)

Country

Intervention and comparator

Total patients (total SP; total AP)

Costs included

Original costs

Adjusted costs (PPP ER, inflation)

Brown (2012)

USA

CORTRAK EAS, conventional blind placement

77 (83; 152)

Tubes, X-rays, intra-hospital transports and staff time

$132 cost saving per episode

£94 cost saving per episode

Gray (2007)

USA

CORTRAK 1 EAS, conventional blind placement

101 (75; 100)

Confirmatory X-rays

$150 cost saving per placement

£115 cost saving per placement

October (2009)

USA

CORTRAK 1 EAS, conventional blind placement

107 (126 SP)

Confirmatory abdominal radiographs

$56 cost saving per placement

£41 cost saving per placement

Sharma (2013)

UK

CORTRAK EAS only

41(48 AP)

Confirmatory X-rays

£83 X-ray related waste per patient

£86 X-ray related waste per patient

Taylor (2010)

UK

CORTRAK 1 EAS, nasogastric tube plus prokinetics

76 (69 SP)

Tubes, prokinetic agents and staff costs

$193 cost saving per treatment course

£143 cost saving per treatment course

Windle (2010)

UK

CORTRAK 1 EAS only

36 (27; 39)

Human resources, consumables like enteral feeding tubes and X-ray films and Nasal-bridle kits

£111 cost per tube insertion attempt with CORTRAK

£122 cost per tube insertion attempt with CORTRAK

Abbreviations: AP, attempted placement; ER, exchange rate; PPP, purchasing power parity; SP, successful placement