Advice

Appendix

Appendix

Contents

Table 2: Overview of the Hetzel et al. (2012) trial

Table 3: Summary of results of the Hetzel et al. (2012) trial

Table 4: Overview of the Jabari et al. (2012) trial

Table 5: Summary of results of the Jabari et al. (2012) trial

Table 6: Overview of the Schumann et al. (2010a) trial

Table 7: Summary of results of the Schumann et al. (2010a) trial

Table 8: Overview of the Hetzel et al. (2004) study

Table 9: Overview of the Schumann et al. (2010b) study

Table 2 Overview of the Hetzel et al. (2012) trial

Study component

Description

Objectives/hypotheses

To evaluate the diagnostic yield and safety of cryobiopsy over conventional sampling (forceps) for endobronchial lesions suspicious for malignancy.

Study design

Multi‑centre single‑blinded randomised controlled trial.

Setting

8 centres in Germany; June 2006 to October 2008.

Inclusion/exclusion criteria

Inclusion:

  • patients with suspected lung lesions based on clinical signs and radiological images

  • patients >18 years old

  • signed informed consent.

Exclusion:

  • patients with bleeding diathesis or who were on anticoagulants

  • patients who had oxygen saturation <90% (under delivery of oxygen at ≤2 l*min‑1)

  • patients with severe underlying cardiac disease.

Primary outcomes

Primary:

  • diagnostic yield of cryobiopsies versus forceps biopsies.

Secondary:

  • duration of biopsy procedure

  • number of samples taken

  • level of difficulty in positioning the probe

  • amount of bleeding.

Statistical methods

Data were analysed by descriptive methods and endpoints were presented separately for each group.

Diagnostic yield was calculated for each biopsy technique as the number of diagnostic procedures divided by the number of non‑diagnostic procedures plus the number of diagnostic procedures.

Based on an explorative test of diagnostic yield, the study was powered at 90% for a significance of p=0.05, for a sample size calculation of n=278 patients.

Primary comparison of diagnostic rate was evaluated by a two‑tailed Chi‑square test. secondary endpoints used Chi‑square and Mann‑Whitney rank tests.

All tests used 5% level of significance.

Patients included

A total of 593 patients were randomised.

Forceps group: n=297, age=65.3±9.9, 69.6% male, height 170.9±8.5 cm, weight 74.3±14.7 kg.

Cryoprobe group: n=296, age=64.8±10.3, 73.4% male, height 170.2±8.9 cm, weight 72.3±15.0 kg.

No significant difference existed between the study groups for the above characteristics.

Results

563 patients were diagnosed with malignant disease. Of these, diagnostic yield was 95.0% for cryobiopsy and 85.1% for standard forceps (p<0.001). Exophytic tumours and submucosal tumours showed better diagnostic yield for cryoprobe than forceps (p=0.003 and p=0.005 respectively).

There was no bleeding in 30.6% of forceps patients and 19.9% of cryoprobe patients (p=0.009). There was mild bleeding (no intervention used) in 51.5% of forceps patients and 61.8% of cryoprobe patients, and severe bleeding (at least 1 intervention for bleeding control was applied) in 17.8% of forceps patients and 18.2% of cryoprobe patients (no p‑value reported). There was no significant difference in the number of bleeding complications needing intervention between the 2 groups (p=0.90).

The proportion of non‑diagnostic results in patients with NSCLC and SCLC was lower after cryobiopsy than after forceps biopsy: NSCLC, 5.5% (95% CI 2.8–9.6) versus 11.8% (95% CI 7.6–17.2), p=0.025; and SCLC, 3.6% (95% CI 0.4–12.3) versus 16.1% (95% CI 8.0–27.7), p=0.024.

Conclusions

Cryobiopsy has a higher diagnostic yield for the diagnosis of endobronchial malignancies when compared with forceps.

Abbreviations: CI, confidence interval; n, number of patients; NSCLC, non‑small cell lung cancer; SCLC, small cell lung cancer.

Table 3 Summary of results from the Hetzel et al. (2012) trial

Forceps

Cryobiopsy

Analysis

Randomised

n=297

n=296

Efficacy

n=297

n=296

Primary outcome: Diagnostic biopsy in patients with malignancy

85.1% (239/297)

95.0% (268/296)

p<0.001

Selected secondary outcomes:

Duration of biopsy procedure

5.25±4.20 min

5.05±4.54 min

p>0.05 min

Number of samples taken

3.45±0.95

3.24±1.16

p<0.009

Level of difficulty in positioning the probe

Not reported

Not reported

p=0.068, in favour of cryobiopsy

Safety

n=297

n=296

–

Patients reporting serious adverse events

Not reported

Not reported

–

Bleeding

  • None 30.6% (91/297)

  • Mild 51.6% (153/297)

  • Severe 17.8% (53/297)

  • None 19.9% (59/296)

  • Mild 61.8% (183/296)

  • Severe 18.2% (54/296)

No bleeding, p=0.009

Abbreviations: n, number of patients.

Table 4 Overview of the Jabari et al. (2012) trial

Study component

Description

Objectives/hypotheses

To evaluate the diagnostic yield and safety of endobronchial biopsies using the flexible cryoprobe.

Study design

Single centre, single‑blinded randomised controlled trial.

Setting

Single‑centre in Iran; April 2009 to June 2011.

Inclusion/exclusion criteria

Inclusion:

  • patients with previously confirmed endobronchial tumour

  • >20 years of age

  • satisfactory respiratory function (oxygen saturation >85% with no oxygen supplementation).

Exclusion:

  • patients with respiratory failure, hypoxia, or respiratory distress

  • patients with a positive history of cardiac or hemodynamic instability

  • abnormality of coagulation.

Primary outcomes

Diagnostic yield, biopsy size.

Statistical methods

Descriptive statistics used for data. The size of the specimen obtained from each sampling method was compared using Friedman's 2‑way Analysis of Variance by Ranks and Wilcoxon signed‑rank test. Cochran's Q test was used to compare diagnostic accuracies.

p‑value <0.05 was considered statistically significant.

Patients included

60 patients; 39 male and 21 female; mean age 56.77±13.3 (range 27–76) years.

Three samples were obtained from each patient.

Results

Diagnosis was made in 57 cases; 54 cases were diagnosed with malignant lesion.

Specimen size obtained (median, range):

  • 5 second cryobiopsy (1.6 cm, 0.9–2.0 cm)

  • 3 second cryobiopsy (0.8 cm, 0.4–1.7 cm)

  • Control arm: forceps biopsy (0.5 cm, 0.1–1.2 cm)

The specimen size obtained with both cryobiopsy techniques compared with forceps biopsy, as well as 5 second compared with 3 second cryobiopsy were statistically significant (p<0.001).

Diagnostic yield was 66.7% for forceps biopsy, 80% for cryobiopsy in 3 seconds and 76.6% for cryobiopsy in 5 seconds, with no significant difference in the methods.

Bleeding complications occurred in 8 cases (13.3%). These were regarded as 'no haemorrhage' in 2 cases (3.3%), 'haemorrhage controlled by normal saline' in 4 cases (6.7%), and 'haemorrhage controlled by diluted adrenaline' in 1 case (1.7%), and there was 1 case of bleeding that needed argon plasma coagulation (1.7%).

Statistical analysis showed that there were no significant differences in cases of bleeding type among the 3 sampling methods (Cochran's Q test; p>0.05).

Conclusions

Diagnostic yield was greater for cryobiopsy when compared with forceps. There was no significant difference in bleeding based on method used.

Table 5 Summary of results from the Jabari et al. (2012) trial

Forceps

Cryobiopsy (3 s)

Cryobiopsy (5 s)

Analysis

Randomised

n=60

n=60

n=60

Efficacy

n=60

n=60

n=60

Primary outcome: diagnostic yield

66.7% (40/60)

80.0% (48/60)

76.7% (46/60)

p>0.05

Selected secondary outcomes:

Size of biopsy sample (median, range)

0.5 cm, 0.1–1.2 cm

0.8 cm, 0.4–1.7 cm

1.6 cm, 0.9–2.0 cm

p<0.001

Safety

n=60

n=60

n=60

Patients reporting serious adverse events

Not reported

Not reported

Not reported

–

Bleeding

Bleeding occurred in 8 cases, not stratified by intervention

Cochran's Q test, p>0.05

Abbreviations: n, number of patients; s, seconds.

Table 6 Overview of the Schumann et al. (2010a) trial

Study component

Description

Objectives/hypotheses

To evaluate the diagnostic yield, feasibility, and safety of endobronchial biopsies using the flexible cryoprobe.

Study design

Single‑centre, prospective, including randomised 2‑arm study.

Setting

Germany, 6 year study (dates not mentioned).

Inclusion/exclusion criteria

Inclusion:

  • exophytic endobronchial tumour (endoscopically visible lesion)

  • signed informed consent form in the first cohort

  • sufficient respiratory function (oxygen saturation >90% with 2 litre of supplementary oxygen).

Exclusion:

  • previous cancer‑specific treatment or endobronchial diagnostic procedure

  • suspected connection of the lesion to large pulmonary blood vessels as seen on chest computed tomography scan

  • thrombocyte count less than 100 G/litre and abnormal plasma clotting

  • age less than 18 years.

Primary outcomes

Diagnostic yield, biopsy area.

Statistical methods

Descriptive statistics were used for quantitative image analysis data. Comparisons of biopsy sample size between FB and CB were made using the Wilcoxon signed‑rank test. Significance was determined by contingency tables and Fisher's exact test.

Patients included

In total, 296 patients were included: 225 males and 71 females; age 63.4±11.8 years, range 20–84. The first 55 patients were randomised consecutively for the sequence of the biopsy procedure (receiving forceps biopsy followed by cryobiopsy or vice versa). All other patients had cryobiopsy only. No demographics were included for the 55 patients.

Results

Diagnostic yield of cryobiopsy was 89.1% (49/55) and 65.5% (36/55) for forceps biopsy (p<0.05). Subgroup analysis showed no influence on these results by the sampling order or by localisation of tumour lesion within the bronchial tree.

The mean total area of each tissue section in the cryobiopsy slides was 10.4 mm2, significantly larger than the mean total area of forceps biopsy (5.2 mm2, p<0.0001). The artefact‑free tissue areas of each slide were significantly greater with cryobiopsy than with forceps biopsy (9.6 mm2 versus 3.6 mm2, p<0.0001).

Bleeding was not reported for the randomised cohort of this trial. Of the 296 patients receiving cryobiopsy, bleeding complications occurred in 15 cases (11 mild, 3 moderate, and 1 severe). Bleeding was graded as mild if ice‑cold water or epinephrine solution was necessary to stop bleeding; moderate if additional techniques (argon plasma coagulator, bronchus blocker) were used; and severe if additional systemic treatment was needed (transfusion of red blood cells, platelets, fresh‑frozen plasma or coagulation factors, fluid resuscitation, use of vasopressors) to stabilise a patient's condition.

Conclusions

Cryobiopsy obtained higher diagnostic yield, larger tissue size and better tissue quality when compared with forceps biopsy.

Abbreviations: CB, cryobiopsy; FB, forceps biopsy; G/litre, giga per litre (109 cells/litre).

Table 7 Summary of results from the Schumann et al. (2010a) trial*

Forceps

Cryobiopsy

Analysis

Randomised

n=55

n=55

Efficacy

n=55

n=55

Primary outcome: Diagnostic yield

65.5% (36/55)

89.1% (49/55)

p<0.05

Selected secondary outcomes:

Mean total area of tissue

10.4 mm2

5.2 mm2

P<0.001

Safety

n/a

n=296

Patients reporting serious adverse events

None reported

None reported

–

Abbreviations: n, number of patients.

*Results from randomised cohort only.

Table 8 Overview of the Hetzel et al. (2004) study

Study component

Description

Objectives/hypotheses

To investigate the effectiveness of CR with a cryoprobe in patients with respiratory tract stenosis caused by exophytic tumours.

Study design

Prospective, single centre.

Setting

Germany, January 2002 to May 2003.

Inclusion/exclusion criteria

Inclusion:

  • High‑grade stenosis in the area of the central respiratory tracts from exophytic tumours with clinical symptoms; high‑grade stenosis was defined as an airway narrowing that could not be passed with an Olympus bronchoscope despite pressure exerted on the bronchoscope

  • stenosis‑closure of a lobe or segment of a bronchus with post‑obstructive pneumonia.

Exclusion:

  • patients with airway stenoses caused by extrinsic compression

  • thrombocyte count less than 100 G/litre

  • abnormal plasma clotting.

Primary outcomes

Success, defined as having no residual stenosis detectable by endoscopy.

Statistical methods

Descriptive statistics used, no analysis.

Patients included

60 patients, age 19–81 years; 23 had complete bronchial obstruction and 37 had high‑grade airway stenosis.

Results

CR was achieved in 37 patients (61%). The procedure was partially successful in 13 patients (22%), meaning that there was residual stenoses that were easily passible with a bronchoscope. CR was unsuccessful in 10 patients (17%).

The procedure lasted 9–81 minutes (41±16 minutes); on average, 13 applications of the cryoprobe were needed during 1 intervention (range 3–31).

Of 57 patients with malignant airway disease, 14 presented again with symptoms of recurrent airway obstruction. Average time between interventions was 18±4 weeks, median 16 weeks, range 10–24 weeks.

After a mean follow‑up time of 25 weeks (range 24–71 weeks), 14 patients were alive without symptoms from airway stenosis. All of these patients had additional chemotherapy, radiation therapy, or both.

No patients died during the procedure. Fifty‑four patients exhibited light bleeding that stopped spontaneously within minutes. The remainder of the patients (n=6) had more intense bleeding (100–300 ml of blood loss), but bleeding was controlled with suction with the flexible bronchoscope and was stopped in all patients.

Conclusions

CR was successful or partially successful in 83% of patients.

Abbreviations: CR, cryorecanalisation; G/litre, giga per litre (109 cells/litre).

Table 9 Overview of the Schumann et al. (2010b) study

Study component

Description

Objectives/hypotheses

To evaluate the efficacy and safety of the ERBECRYO flexible cryoprobe (length and diameter not stated) for immediate tumour ablation in patients with exophytic endobronchial or tracheal tumour obstruction.

Study design

Single‑centre, retrospective study.

Setting

Germany, February 2001 to September 2007.

Inclusion/exclusion criteria

Inclusion:

  • patients who had bronchoscopic intervention with a flexible cryoprobe because of an exophytic tumour of the trachea or bronchi with subsequent airway stenosis.

Exclusion:

  • patients with platelet count<100 G/litre

  • abnormal coagulation parameters (partial thromboplastin time>40 seconds and prothrombin ratio <70%).

Primary outcomes

Success rate of CR, defined as successful if the endobronchial tumour mass could be ablated so that either drainage of secretion or reduced airway stenosis led to an improvement in the patient's condition and symptoms. CR was not successful if the primary goal of the intervention (i.e. target bronchus reopening) could not be reached.

Safety: severity of bleeding, need for a rescue operation, or prolonged mechanical ventilation >2 hours after the intervention.

Statistical methods

Descriptive statistics used for patient population. No statistical analysis.

Patients included

225 patients, 156 men (69.3%); mean age 63.9±12.9 years (range 19–83 years); 193 (85.8%) inpatients.

Results

197/225 (87.6%) had malignant airway stenosis; 147 (74.6%) had lung cancer.

CR was successful in 205/225 patients (91.1%). In 20 patients (8.9%), the intervention was not successful and did not reopen the obstructed bronchus. The main reason for unsuccessful interventions was longer length of the complete bronchial obstruction.

Most of the interventions were done only with the flexible technique (194 patients). Additional techniques were used in 48 patients: 37 (16.4%) had APC and 11 (4.9%) had stent implantation.

Bleeding complications occurred in 27 (12%) patients. Bleeding was defined as mild if the application of cold NaCl 0.9% solution (2–4ºC) or epinephrine solution (1 mg/10 ml saline water) applied topically was sufficient to stop bleeding (9 patients experienced mild bleeding). Bleeding was defined as moderate if additional techniques (APC or a bronchus blocker) had to be used (18 patients experienced moderate bleeding).

Conclusions

CR was successful in most patients.

Abbreviations: APC, argon plasma coagulation; CR, cryorecanalisation; G/litre, giga per litre (109 cells/litre); NaCl, sodium chloride.