Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Graves (2019)

Type of study:

RCT

Setting and country:

Single-center; surgeries between May 23, 2016 and March 16, 2018; USA

Funding and conflicts of interest:

“This work was funded by the National Institutes of Health, NIH Grant 5R03HD090617–02.”

Conflicts of interest not described in publication.

Inclusion criteria:

Only otherwise healthy patients with pectus excavatum deformity

Exclusion criteria:

• age < 13 years at time of procedure;
• chronic use of pain medication preoperatively;
• pectus carinatum, Poland's syndrome, or any chest wall anomaly other than pectus excavatum;
• previous repair of pectus excavatum by any technique;
• previous thoracic surgery;
• congenital heart disease;
• bleeding dyscrasia;
• major anesthetic risk factors or history of previous problem with anesthesia;
• pregnancy;
• non-English speaker.

N total at baseline:

Total: 20

Intervention: 10

Control: 10

Important prognostic factors²:

Ag, median (range):

I: 20.9 (14–31)

C: 16.1 (13–21)

P=0.03

Sex:

I: 9/10 (90%) M

C: 8/10 (80%) M

Height, mean (cm)

I: 176.2

C: 175.8

P=0.94

Weight, mean (kg)

I: 62.9

C: 61.2

P=0.90

ASA Class

ASA I

I: 5/10 (50%)

C: 5/10 (50%)

ASA II

I: 5/10 (50%)

C: 5/10 (50%)

Groups not comparable at baseline; intervention group was older (MD 3.8y).

Intercostal nerve cryoanalgesie - during Nuss procedure

Patients randomized to cryoanalgesia were intubated with a doublelumen endotracheal tube for better lung isolation and exposure of the posterolateral course of the intercostal nerves. Patients were treated with the AtriCure cryosurgical system (AtriCure, Inc., West Chester, Ohio) during the Nuss procedure. All wounds were infiltrated with 0.25% Marcaine prior to incision. An additional 5 mm incision was made on each side in the midclavicular line low in the chest for the thoracoscope, and the cryoprobe was inserted through the bar insertion wound. The cryoprobe was directly applied to the intercostal nerve, posterolateral to the chest wall incisions, under thoracoscopic visualization at the level of the incision and 2 interspaces above and below, bilaterally. Cryoablation was performed by applying the cryoprobe to the intercostal nerves for 2 min to achieve a temperature of −60 °C. Each nerve received one cycle of cryoablation.

See column of control procedure for details on Nuss procedure.

Post-operative care:

Postoperative care and analgesic protocol were identical for all patients with the exception of continuous epidural infusion (no bolus or patient-controlled epidural analgesia). All patients were admitted to the pediatric inpatient unit under the pediatric surgery service. The study team, in conjunction with the pediatric pain service, developed a multimodal postoperative pain regimen (Appendix A), which was used to guide medication prescription and dosing by the pediatric surgical service for patients in both treatment groups. Local field block was performed intraoperatively at all incision sites. All patients received a hydromorphone PCA, standing intravenous acetaminophen, and 48 h of standing ketorolac immediately postoperatively. Once patients were tolerating a diet, they were converted to oral oxycodone, acetaminophen, and ibuprofen. Per institutional protocol, all epidurals were weaned at the discretion of the pediatric pain management team in consultation with the primary surgical team. Both managing teams assessed patients twice daily for readiness to wean and/or discontinue the epidural. Narcotic usage was recorded and tracked through the hospital's electronic medical record.

Upon discharge, patients were sent home with acetaminophen, ibuprofen and oxycodone and were instructed to keep a medication log.

Epidural analgesia - during Nuss procedure

Patients randomized to epidural analgesia were intubated with a single-lumen endotracheal tube. A thoracic epidural catheter was placed by the pediatric anesthesia team immediately prior to the Nuss procedure. Catheters were placed using sterile technique at approximately T5–6 or T6–7 interspaces; procedure specifics were left to the discretion of the attending anesthesiologist. After placement, epidural infusion was begun with 0.1% ropivacaine and 2 μg/cc fentanyl. No bolus doses were administered.

Nuss procedure: All patients underwent the Nuss procedure, performed by a single attending pediatric surgeon with extensive experience performing the procedure. Specifics of the procedure (i.e., one vs. two bar implants, location of fixation, number of stabilizers) were left to the discretion of the operating surgeon based on individual patient factors. However, the same general repair protocol was followed for all patients: small incisions were made on each side of the chest wall, and a subcutaneous tunnel was created for Nuss bar insertion just medial to the pectus ridge. Separate incisions were made more laterally for insertion of the thoracoscope. Under thoracoscopic guidance, the retrosternal space was developed and a Lorenz passer was advanced across the mediastinum. The passer was exchanged for a Nuss bar, which was bent to a custom configuration for each patient. The bar was introduced with convexity facing posteriorly, then flipped inside the patient, so that the convexity was

anterior, which forced the sternum anteriorly and corrected the pectus excavatum. The bar was secured to the ribs with bilateral stabilizers.

See column of intervention procedure for details on postoperative care.

Length of follow-up:

2 weeks, 1 months, 3 months and 1 year from date of surgery

Loss-to-follow-up:

Intervention:

N = 1 (10%); no reason reported

Control:

N = 0 (0%)

Incomplete outcome data:

2 patients of the intervention group (20%) did not complete the 1-year follow-up

(“Patients undergoing cryoanalgesia received postoperative follow-up to one year with the exception of one patient who had not yet reached the one year mark, and one patient who was lost to follow-up after seven months.”)

Unclear why overall pain score is present for control group and missing for the intervention group at day 5.

Postoperative pain

Overall Pain Score*, mean (rate overall pain on scale of 1 to 10, where 10 is max)

Acute (up to 7 days after surgery)

Day 1

  I: 3.1

  C: 3.0

  P=0.88

Day 3

  I: 2.8

  C: 2.9

  P=0.78

Chronic (3 months after surgery)

3 Months

  I: 1.3

  C: 1.1

  P=0.34

1 Year

  I: 1.3

  C: 1.1

  P=0.34

Other follow-up

2 Weeks

  I: 2.2

  C: 2.1

  P=0.91

1 Month

  I: 2.5

  C: 1.9

  P=0.58

Rescue medication (including opioids and opioid-related side effects)

Opioid analgesia during postoperative stay:

I: 268 mg (range 150–386 mg; Std Dev 165.2 mg)

C: 684 mg (range 547–821 mg; Std Dev 191.8 mg)

MD -416 mg; p = 0.0001)

Decreased requirement not attributable to shorter length of postoperative stay based on comparison of daily opiate requirements.

Day 1: 52% less opioid

Day 2: 76% less opioid

Day 3: 82% less on POD 3

p-values <0.01

Ketorolac (NSAID):

MD -89.3 mg, p = 0.02)

Acetaminophen (paracetamol):

MD -848.1 mg, p = 0.01

Neuropathic pain

Neuropathic Pain Score*, neuropathic-type pain, described as “burning”, “electrical” or “tingling” sensationsmean;

2 Weeks

  I: 3.0

  C: 1.9

  P=0.32

1 Month

  I: 1.7

  C: 1.3

  P=0.37

3 Months

  I: 1.9

  C: 2.1

  P=0.80

1 Year

  I: 2.4

  C: 1.9

  P=0.62

Anterior chest numbness (%)

2 Weeks

I: 10/10 (100%)

C: 2/10 (20%)

1 Month

I: 10/10 (100%)

C: 2/10 (20%)

3 Months

I: 6/10 (60%)

C: 0/10 (0%)

1 Year

I: 0/10 (0%)

C: 0/10 (0%)

Other adverse events

Complications, patients with complication

I: 0/20

C: 2/20

“There were no complications in the cryoablation group. In the epidural group, one patient required insertion of a chest pigtail drain for symptomatic pneumothorax on POD 0, and another patient had a urinary catheter inserted for postoperative urinary retention.”

Length of hospital stay

Defined as discharge on postoperative day, median:

I: 3 (range: POD 2–4),

C: 5 (range: POD 4–6

All patients in intervention group were discharged by POD 4; 4 patients of control group were discharged by POD 4 and the other 6 were discharged on POD 5 or 6.

Also reported:

Room time, operative time, anterior chest numbness (2w, 1m, 3m, 1y), number of successful procedures, complication rate.

Definitions:

* Patients were asked to rate the overall pain and neuropathic pain on a scale of 0 to 10, with 10 being the maximum pain score.

Remarks:

-some concerns regarding risk of bias

Authors conclusion:

Intercostal nerve cryoablation during the Nuss procedure decreases hospital length of stay and opiate requirement versus thoracic epidural analgesia, while offering equivalent pain control.

Ju (2008)

Type of study:

RCT

Setting and country: single center, Peking University People’s Hospital, Beijing, China

Funding and conflicts of interest:

“This study is

supported by grants from R&D Foundation of PKUPH

(C906).”

Conflicts of interest not reported in publication.

Inclusion criteria:

• patients scheduled for posterolateral thoracotomy

for lung or esophagus diseases

• physical status I or II

Exclusion criteria:

• coagulopathy
• skin infection at the site of intended epidural puncture,
• preoperative respiratory dysfunction
• analgesic ingestion history for chronic pain

N total at baseline:

I: 53

C: 61

Included in study:

I: 53

C: 54 (7 patients excluded because of epidural catheter dislocation within two postoperative days)

Important prognostic factors²:

age ± SD:

I: 61.41 ± 11.78

C: 61.80 ± 13.78

Sex, M:

I: 38/53 (72%)

C: 41/54 (76%)

Demographics of groups comparable at baseline.

Operation characteristics:

Operation duration (min)

I: 249.77 ± 79.85

C: 232.21 ± 68.18

Propofol doses during operation

I: 1441.95 ± 669.30

C: 1037.18 ± 373.01

Fentanyl doses during operation

I: 0.36 ± 0.10

C: 0.20 ± 0.07

Operation types (P/E; pulmonary operation/esophagus operation)

I: 25/28

C: 28/26

Doses of propofol and fentanyl during operation higher in intervention group.

Intercostal nerve cryoanalgesia

- during posterolateral thoracotomy for lung or esophagus diseases

Lateral thoracotomies

(either left or right) were chosen and one-lung ventilation was required for all patients.

After arriving at the operating room, the patients were

premedicated with 0.3 mg scopolamine and 0.02 mg/kg midazolam intraveneously (IV), and were given 10 ml/

kg/h Ringer’s lactate solution (Baxter, USA). The

patients were monitored with electrocardiography, pulse oximetry, Bispectral index (BIS, Pillip Medical System, USA) and invasive blood pressure measurements.

Epidural catheters were inserted through the T6-7 or T7-8 subcutaneously (Group C) preoperatively. In order to make the patients blinded to the analgesic method, subcutaneous infusion catheters

were inserted at upper back (T7-8 level) in Group C.

Anesthesia was induced with 3 lg/kg fentanyl, 2 mg/kg propofol and 0.6 mg/kg rocuronium. A double lumen tube was applied for one-lung ventilation. The patients

were also monitored with central venous pressure after the induction of anesthesia administration. The propofol dose was titrated to maintain BIS values at a range of 50–

55. The neuromuscular blockade was maintained with 0.3–0.6 mg/kg/h atracurium.

The intra-operative analgesia was maintained with incremental doses of fentanyl citrate in Group C, but fentanyl administration was ceased 1 h before the end of surgery to eliminate its analgesic effect during the early postoperative period.

All patients were extubated in operating room and transferred to the postanesthesia care unit (PACU).

For the patients in group C, before closure of the thorax, the intercostal nerves (one at the level of the incision, one cranial and one caudal) were identified and exposed by peeling off the parietal pleura. The Kooland cryoprobe (administering CO2 as the cooling agent, JP-1, Kooland, China) was placed on each nerve, under

direct vision. Each nerve received a 90 s application of cold (-70 °C). After the operation, these patients

received patient-controlled analgesia through the subcutaneous catheter (s.c., 1 mg/ml morphine, 2 ml PCA and a lock-out interval of 30 min, without basal infusion) for 72 h postoperatively.

This PCA pump served as rescue medication, and patients were asked to use PCA, if their numeral rating scores (NRS) were > 4.

Epidural analgesia

- during posterolateral thoracotomy for lung or esophagus diseases

Lateral thoracotomies

(either left or right) were chosen and one-lung ventilation was required for all patients.

After arriving at the operating room, the patients were

premedicated with 0.3 mg scopolamine and 0.02 mg/kg midazolam intraveneously (IV), and were given 10 ml/

kg/h Ringer’s lactate solution (Baxter, USA). The

patients were monitored with electrocardiography, pulse oximetry, Bispectral index (BIS, Pillip Medical System, USA) and invasive blood pressure measurements.

Epidural catheters were inserted through the T6-7 or T7-8 into the epidural space (Group E) preoperatively. In order to make the patients blinded to the analgesic method, subcutaneous infusion catheters

were inserted at upper back (T7-8 level) in Group C.

Anesthesia was induced with 3 lg/kg fentanyl, 2 mg/kg propofol and 0.6 mg/kg rocuronium. A double lumen tube was applied for one-lung ventilation. The patients

were also monitored with central venous pressure after the induction of anesthesia administration. The propofol dose was titrated to maintain BIS values at a range of 50–

55. The neuromuscular blockade was maintained with 0.3–0.6 mg/kg/h atracurium.

In the Group E, the epidural analgesia was initiated before the incision and maintained with 5–10 ml/h 0.5% ropivacaine

during the operation. IV fentanyl was not administered

after the induction of anaesthesia for the patients in group E.

All patients were extubated in operating room and transferred to the postanesthesia care unit (PACU).

During the first 72 h following the operation, the patients in group E used the patient-controlled epidural analgesia (PCEA) (0.125% bupivacaine + 0.05 mg/ml morphine + 0.02 mg/ml droperidol, basal rate was 3 ml/h, 3 ml patient-controlled analgesia (PCA) at a lock time of 15 min).

This PCA pump served as rescue medication, and the patients were asked to use the PCA, if their numeral rating scores (NRS) were > 4.

Length of follow-up:

12 months;

Postoperative pain: 8h

Use of analgesic in intervention group: 3 postoperative days

Telephone follow-up for pain scores: 1M, 3M, 6M, 12M

Loss-to-follow-up:

I: 14/53 (26.4%)

Reasons (describe)

4 loss of connection, 10 died

C: 16/54 (29.6%)

Reasons: 4 loss of connection, 12 died

Incomplete outcome data:

for follow-up pain measures:

I: up to 26.4%

C: up to 29.6%

For reasons, see above.

Postoperative pain

Acute

No data provided; “The numeral rating scales (NRS) at rest were around 1 point, and 3 points when coughing. No

matter at rest or on coughing, NRS were slightly lower among the patients in group E than those in group C. However, no statistically significant differences were found between the 2 groups with respect to NRS pain scores at rest or on motion within 3 days following surgery.”

Chronic

Defined as any kind of pain experienced, spontaneous or touch evoked, intermittent or persistent; results also reported for 1M and 6M; also reported: interference with daily life)

Incidence - Chronic pain

3M

I: 33/48 (68.8)

C: 31/50 (62.0)

12M

I: 22/39 (56.4)

C: 16/38 (42.1)

Intensity - Moderate to severe pain

3M

I: 11/48 (22.9%)

C: 4/50 (8.0%)

12M

I: 9/39 (23.1%)

C: 2/38 (5.3%)

The other patients experienced no pain or mild pain.

Rescue medication

(including opioids and

opioid-related side effects)

Side effect of epidural morphine administration for postoperative pain relief

-nausea, vomiting, and sedation: similar in the two groups; no data provided

-most common: pruritus

-incidence of pruritus in group E was significantly higher in epidural group than control group (p < 0.05); no data provided.

Rescue medication

Near 97% of the patients in group C received morphine subcutaneously as rescued medication (average dose (23.3 +/- 22.2) mg/3 d.

Neuropathic pain

Incidence - Allodynia-like pain; also reported for 1M

3M

I: 9/48 (18.8%)

C: 3/50 (6.0%)

6M

I: 7/43 (16.3%)

C: 1/48 (2.1%)

12M

I: 6/39 (15.4%)

C: 0/38 (0%)

Other adverse events

Nausea, vomiting, sedation and pruritus

No data reported; “The incidences of nausea, vomiting, and sedation

were similar in the two groups (see Fig. 1). Pruritus

was the most common side effect of epidural morphine

administration for postoperative pain relief. The incidence of pruritus in group E was significantly higher

than that in group C (p < 0.05).”

Length of hospital stay

Postoperative hospitalisation (days), mean±SD

I: 12.19 ± 5.12

C: 12.36 ± 5.08

Drainage (days)

I: 8.19 ± 4.13

C: 7.74 ± 3.68

Chest tube extubation

No significant differences were observed

Also reported: deaths before discharge, deaths within 12 months

Definitions:

-

Remarks:

-high risk of bias

Authors conclusion:

Although intercostal nerve cryoanalgesia combined

with small dose subcutaneous morphine can provide effective postoperative pain control, it should not be recommended for post-thoracotomy pain control until the mechanism of cryoanalgesia-induced neuropathic pain is revealed and the effective prevention is developed.

Mustola (2011)

Type of study:

RCT

Setting and country:

South Carelia Central Hospital, Finland

Funding and conflicts of interest: not reported in publication

Inclusion criteria:

• scheduled for elective posterolateral thoracotomy

Exclusion criteria:

• Requirement of

chronic opioid therapy

• moderate or

severe pain—verbal pain scale (VPS) of 2 or more —at the operation site before the operation

N total at baseline:

Intervention: 21

Control: 21

Important prognostic factors²:

age ± SD:

I: 64.6 ± 11.1

C: 59.5 ± 14.9

Sex:

I: 13/21 (61.9%) M

C: 15/21 (71.4%) M

ASA score

Score I

I: 0, C: 2

Score II

I: 10, C: 9

Score III

I: 11, C: 10

Groups comparable at baseline.

Intercostal nerve cryoanalgesia + thoracic epidural analgesia

- during elective posterolateral thoracotomy

For details on surgery and epidural analgesia, see column of control intervention.

Cryoanalgesia:

At the end of the operation, the surgeon gave 60

seconds of cryoanalgesia at –70°C (140 Cryo Unit, Spembly Medical, Hampshire, UK) to 3 intercostal nerves: 1 at

the level of the incision, 1 caudal, and 1 cranial. Cryoanalgesia was performed on the intercostal nerves about 10

cm from the nerve root. The investigating anesthetists

(S.M., J.L.) and patients were blinded to whether cryoanalgesie was performed.

Intracostal sutures, USP 0 Maxon (Davis & Geck,

Davis, NJ) through drilled (Mini Air Drill, Synthes, Oberdorf, Switzerland) 2-mm holes in adjacent ribs, were used during chest closure to avoid the impingement of intercostal

nerves (Fig 1).

For postanesthesia care, see column of control intervention

Thoracic epidural analgesia

- during elective posterolateral thoracotomy

The electrocardiogram, noninvasive or invasive blood pressure, heart rate, end-tidal carbon dioxide, end-tidal isoflurane concentration, peripheral oxygen saturation, and train-of-four were monitored using an S/5 monitor (GE Healthcare Finland Oy, Helsinki, Finland). Thoracic epidural catheters were inserted through the T5 to T6, T6

to T7, or T7 to T8 interspaces before induction of anesthesia.

Patients were premedicated with diazepam (0.1 mg/kg,

adjusted to the nearest 2.5 mg) and received fentanyl (0.1 to 0.2 mg) and thiopental (3 to 5 mg/kg) for induction of anesthesia. Rocuronium was used for neuromuscular blockade. One-lung ventilation was used during the operation, and anesthesia was maintained with isoflurane (1% to 3% in 40% to 100% oxygen) and fentanyl boluses (0.05 to 0.1 mg) as needed.

Before the end of anesthesia, a 0.1-mg fentanyl bolus with physiologic saline (5 mL) was given to the epidural catheter.

In the postanesthesia care unit, an epidural infusion of

ropivacaine (1 mg/mL) plus fentanyl (5 µg/mL) was started at a rate was of 3 to 10 mL/h, and 4-mL boluses were given as needed. Epidural infusion was continued for 3 days on

the ward, and the rate was adjusted according to VAS (0 to 100 mm), with the goal of maintaining the VAS at 30 or less at rest and at 40 or less while walking.

The epidural infusion was stopped after 3 days, and

pain was treated with oral strong (oxycodone) or weak

(tramadol or codeine) opioids combined with nonsteroidal anti-inflammatory drugs or acetaminophen, as

needed.

Length of follow-up:

Up to 6 months;

Acute pain: during first 12 hours

Pain: 8W, 6M

Loss-to-follow-up:

Intervention: 1/21 (4.8%)

Control: 1/21 (4.8%)

Reason: incomplete data, for which excluded from analysis

Incomplete outcome data:

Intervention: 1/21 (4.8%)

Control: 1/21 (4.8%)

All patients visited the local pain

clinic 8 weeks after operation, but 3 patients in the EC

group and 2 in E group did not visit the local pain clinic

at 6 months and were interviewed by telephone.

Postoperative pain

Acute pain, VAS-score (scale 0 to 100 mm)

1 day – at rest

  I: 13.8±13.9 (95% CI 7.7 to   

  19.9)

  C: 8.6±12.3 (95% CI 3.2 to

  14.0),

  p= 0.21

1 day – in movement

  I: 45.9±21.5 (95% CI 36.2 to

  55.7)

  C: 31.7±17.6 (95% CI 23.6 to 39.7)

  p=0.052

2 days – at rest

  I: 12.8±13.7 (95% CI 6.8 to

  18.8)

  C: 3.5±8.7 (95%CI -0.3 to 7.2)

  P= 0.057

2 days – in movement

  I: 36.0±20.0 (95% CI 26.8 to 

  45.1)

  C: 29.5±22.4 (95% CI 19.3 to

  39.7)

  P=0.332

3 days – at rest

  I: 8.1±12.1 (95% CI 2.8 to

  13.4)

  C: 5.3±11.1(95% CI  0.5 to

  10.1)

  P= 0.54

3 days – in movement

  I: 32.6±22.1 (95% CI 22.6 to  

  42.7)

  C: 27.9±19.4 (95% CI 19.0 to

  36.7)

  P= 0.597

Chronic pain, VAS score (scale 0 to 100 mm; scores also available for 1W, 4W, 8W)

6 months – at rest

  I: 6.3±11.5 (95% CI 1.3 to

  11.3)

  C: 2.2±6.5 (95% CI -0.6 to 5.0)

  P=0.291

6 months – in movement

  I: 19.3±27.2 (95% CI 6.9 to

  31.7)

  C: 7.6±16.1 (95% CI 0.3 to

  14.9)

  p=0.281

Pain scores were also available for 1W, 4W, 8W and for VPS (verbal pain scores) for all these moments of follow-up.

Rescue medication

(including opioids and

opioid-related side effects)

Epidural infusion rates during 3 postoperative day in the ward

I: 4.7±0.6

C: 5.1±0.5

mL/t)

Number of boluses during 3 postoperative day in the ward

I: 6.2±4.9

C: 5.8±4.7).

Oxycodone requirements after epidural infusion was stopped I: 23.1±27.1 mg/3 d

C: 38.4±66.9 mg/3 d

Requirement for medication at home (tramadol (50 mg) or codeine (30 mg)

and acetaminophen (500 mg), whereas others needed

only nonsteroidal anti-inflammatory drugs or acetaminophen)

I: 14/20

C: 15/20

Neuropathic pain

(Hypoesthesia

8W

I: 20/20

C: 10/20

6M

I: 8/20

C: 6/20)

Allodynia

8W

I: 9/20

C:  3/20

6M

I: 5/20

C: 2/20

Hyperalgesia

8W

I: 2/20

C: 0/20

6M

I: 1/20

C: 0/20

Dysesthesia

8W

I: 1/20

C: 1/20

6M

I: 0/20

C: 1/20

Other adverse events

Not reported

Length of hospital stay

Not reported

Definitions:

-

Remarks:

-some concerns regarding risk of bias

Authors conclusion:

In conclusion, the findings of the present study are in

accordance with previous studies where intercostal nerve cryoanalgesia increased the risk of long-term and neuropathic-type pain after thoracotomy. Furthermore, thoracic epidural analgesia combined with the drilled holes technique can decrease the risk of chronic postthoracotomy pain. Cryoanalgesia is not recommended for the

treatment of postthoracotomy pain.

Sepsas (2013)

Type of study:

RCT

Setting and country: Sotiria’ General Hospital for Chest Diseases, Greece

Funding and conflicts of interest:

funding is not reported. No conflicts of interest declared.

Inclusion criteria:

• lung cancer patients
• undergoing thoracotomy and pulmonary resection (lobectomy, bilobectomy, pneumonectomy)

Exclusion criteria:

• ASA score ≥4
• age >75
• BMI >35
• history of other malignancy,
• sleep apnoea
• refusal to
• give informed consent

N total at baseline:

Intervention: 25

Control: 25

Important prognostic factors²:

age ± SD:

I: 64.4 ± 7.1

C: 64.8 ± 8.6

Sex, M:

I: 21/25 (84%)

C: 21/25 (84%)

ASA (1/2/3)

I:  8/13/4

C: 8/16/1

Groups comparable at baseline.

Cryoanalgesia, combined with intravenous

patient-controlled analgesia (IVPCA)

– during thoracotomy and pulmonary resection

(lobectomy, bilobectomy, pneumonectomy)

For information about the IVPCA, see column of control group.

Cryoanalgesia

Group A patients were submitted to one session of cryoanalgesia (−40°C) for 60 s, under direct vision —up to 10 cm from intercostal nerve outgrowth and definitely before its bifurcation—at the thoracotomy level, as well as one level above and two levels below. Cryoanalgesia sessions were performed by the same surgeon using a Spembly 142 Cryo Unit

(Spembly Medical, Andover, UK).

Postoperative care

Immediately after the patient recovered consciousness, time

to extubation was recorded and the intensity of postoperative

pain was evaluated using the 1-unit verbal analogue scale. When

pain was ≥3 at rest or ≥5 on the verbal analogue scale during coughing, additional intravenous morphine (2.5 mg) was administered

at 10 min intervals until pain was restored to these

levels. Subsequently the patient was connected to a RythmicTM pump (Micrel Medical Devices, Gerakas, Greece) for patient controlled intravenous analgesia (PCA), containing a 2 mg/ml morphine solution with 1 mg dosage and 10 min lockout.

Intravenous patient-controlled analgesia (IVPCA) alone

- during thoracotomy and pulmonary resection

(lobectomy, bilobectomy, pneumonectomy)

IVPCA

General anaesthesia was provided throughout by the same team of anaesthesiologists, using a specific technique: propofol

and remifentanil for induction to general anaesthesia, a nondepolarizing

neuromuscular blocker, single lung ventilation with

the use of a double lumen endobronchial tube or endobronchial blocker, remifentanil for intraoperative analgesia and sevoflurane (without N2O) for maintenance of the anaesthesia. In order to

achieve postoperative analgesia, morphine 0.3 mg/kg, tenoxicam

16 mg and paracetamol 2 g were administered intravenously approx.

30 min before the end of the surgical procedure.

The procedure was performed through a posterolateral thoracotomy

via the upper rim of the 5th rib. Pericostal sutures were

applied around both ribs.

For information about postoperative care, see column of intervention group.

Length of follow-up:

60 days;

Loss-to-follow-up:

Intervention:

0 (0%)

Control:

0 (0%)

Incomplete outcome data:

n.a.

Postoperative pain

Acute pain

Postoperative pain at rest, VAS scale, mean±SD (max recorded value, min. recorded value)

6 hours postop.

  I: 0.9±0.7 (2, 0), n= 25

  C: 3.00±1.00 (5,2), n= 25   

  p<10−4

24 hours postop.

  I: 0.6±0.6 (2, 0), n= 25

  C: 2.45±1.2 (6, 1), n= 25

  P <10−4

48 hours postop.

  I: 0.5±0.5 (1, 0), n= 25

  C:  2.1±0.95 (5, 1), n= 25

  P <10−4

72 hours postop.

  I: 0.4±0.5 (1, 0), n=25

  C:  1.6±0.8 (4, 1), n= 25

  P <10−4

Chronic pain

30 days postop.

  I: 0.0±0.0 (0, 0), n=25

  C: 1.0±1.1 (3, 0), n= 25  

  p<10−4

60 days postop.

  I:  0.0±0.0 (0, 0), n= 25

  C: 0.25±0.45 (1, 0), n= 25

  P=0.01

Pain scores also reported for 12, 18, 36, 60, 84 and 96 hours and 5, 6, 7 and 14 days.

Rescue medication

(including opioids and

opioid-related side effects)

Morphine dosage administered (mg) during the immediate postoperative period, mean±SD

Hour 0

  I: 0.0±0.0

  C: 1.25±2.6

  P= 0.03

Hours 24–36

  I:  1.55±0.9

  C: 10.1±2.6

  P<10−4

Hours 48–60

  I: 1.25±0.8

  C: 9.1±2.0

  P <10−4

Hours 72–84

  I: 0.15±0.3

  C: 9.0±1.9

  P <10−4

Dosages also reported for 0-6, 6-12, 12-18, 18-24, 36-48, 60-72, 84-96 hours post-op.

Analgesic regimens administered per patient during the late postoperative period (days 5–60); defined as oral analgesics administered and total 24 h dosage, reported as 0/TR/CD/PC/CL*;

Day 7 postop.

  I: 14/0/6/11/0

  C: 0/6/19/19/20

  P <10−4

Day 14 postop.

  I: 17/0/1/7/0

  C: 0/8/14/25/2

  P <10−4

Day 60 postop.

  I: 25/0/0/0/0

  C: 7/3/15/15/0

  P <10−4

Neuropathic pain

Acute; number of patients reported non/few/moderate dysesthesia

0 hours postop.

  I: 11/10/4

  C: 4/18/3

  P=0.28 (SNS)

24 hours postop.

  I: 14/8/3

  C: 5/19/1

  P=0.004

48 hours postop.

  I: 16/8/1

  C: 4/20/1

  P=0.004

72 hours postop.

  I: 17/7/1

  C: 1/24/0

  P=10-4

Also reported for 6, 12,18, 36, 60, 84 and 96 hours postop.

Chronic number of patients reported non/few/moderate dysesthesia

7 days postop.

  I: 17/8/0

  C: 8/17/0

  P=0.01

14 days postop.

  I: 18/7/0

  C: 9/16/0

  P=0.01

60 days postop.

  I: 19/6/0

  C: 9/15/1

  P=0.02

Also reported for 5, 6 and 30 days postop.

Other adverse events

Epigastric distension and back pain

No data provided; “Group B patients demonstrated significantly increased frequency of epigastric distension and back pain, compared with Group A

(P < 0.0001) at all assessments performed, from day 5 up until

two months after the surgical procedure.”

Nausea and vomiting

Nausea:

I:  0/25 (0%)

C: 5/25 (20%)

“Nausea was more frequent in

Group B patients (20%) compared with Group A patients (0%),

between hours 12 and 18 postop; however, vomiting was equally present in both groups. No patient demonstrated nausea

or vomiting after hours 24 and 18, respectively.”

Length of hospital stay

Not reported

Also reported:

Frequency of nausea and vomiting; mean arterial blood pH; mean heart rate, forced expiratory volume in 1s

(FEV1) and forced expiratory vital capacity (FVC)

Definitions:

0/TR/CD/PC/CL*, implying 0 = none; TR = tramadol 200 mg; CD = codeine 90 mg; PC = paracetamol 1.5 g;

CL = celecoxib 400 mg

Remarks:

-some concerns regarding risk of bias

Authors conclusion:

Post-thoracotomy analgesia remains a multifactorial, ambiguous and intractable issue. Intraoperative intercostal cryoanalgesia

seems to be an auxiliary, cost-effective and efficient method, capable of controlling post-thoracotomy pain and restoring respiratory

function in the majority of cases. Furthermore, cryoanalgesie reduces the frequency of unpleasant postoperative events such as nausea and dysesthesia.

Study reference

(first author, publication year)

Was the allocation sequence adequately generated? ^a

Definitely yes

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?^b

Definitely yes

Probably yes

Probably no

Definitely no

Blinding: Was knowledge of the allocated

interventions adequately prevented?^c

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?^d

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?^e

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?^f

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure^g

LOW

Some concerns

HIGH

Graves (2019)

Definitely yes

Reason: “Before enrollment, the numbers 1 to 20 were randomly assigned to cryoanalgesia or thoracic epidural groups in a 1:1 ratio using a computerized random sequence generator (http://www.random.org).”

Definitely yes

Reason: “The assignments were placed in closed envelopes by a third party, uninvolved with the study, with the integers on the outside of the envelopes. As patients were enrolled in the study, they were consecutively assigned their patient number in order of enrollment. The envelope associated with each patient number was opened in the preoperative area just prior to time of surgery, revealing the patienťs group assignment.”

Definitely no

Reason: not possible to blind for intervention.

“Patients and physicians were blinded to study arm until the envelope was opened in the preoperative waiting area.” + “Unfortunately, blinding to the intervention was not feasible in this study either for the surgeon or for the patient who awoke with or without an epidural catheter. Additionally, we felt that to ensure patient safety and to optimize individualized analgesia regimens, it was critical that the healthcare team be aware of which treatment patients received.”

Definitely no

Reason: no lost to follow-up reported and flow chart shows complete data for all patients. However, in text described that 2 patients of the intervention group (20%) did not complete the 1-year follow-up (1 did not reach 1-year mark and 1 was lost to FU).

Probably no

Reason: multiple  outcomes not announced in study protocol; costs analysis announced but not performed and reported.

ClinicalTrials.gov Identifier: NCT02721017

Probably yes

Reason:

adherence to ITT protocol; no problems regarding sponsor, premature ending etc. reported

Some concerns (all outcomes)

Reason: no blinding, inconsistent reported of number of patients per outcome measure (2 patients (20%) did not complete follow-up; not shown in flow chart); multiple outcomes different from outcomes announced study protocol.

Ju (2008)

Randomization method not reported

Reason: no details provided. “Patients were stratified by disease sites (lung or esophagus), and blinded randomized

to receive either […].”

Probably no

Reason: no details provided

Probably no

Reason: no procedure described for in-hospital blinding of staff or data-analysts. Most of the outcomes possibly susceptible for bias.

Patients: “In order to make the patients blinded to the analgesic method, subcutaneous infusion catheters were inserted at upper back (T7-8 level) in Group C.”

Assessor: only reported for telephone follow-up interviews; “One, three, six, and twelve months after surgery, an investigator, who was blinded to the

postoperative pain management, interviewed patients

by telephone, using a standard questionnaire.”

Definitely no

Reason: high number of loss-to-follow-up and incomplete data (loss of connection, death; up to 26-29%)

Probably yes

Reason: no protocol or study registration reported. Multiple results described instead of data reported. Relevant outcome measures reported with plausible length of follow-up.

Definitely no

Reason: high number of lost to follow-up, no imputation of data or comparison of patients that were kept included with patients that were lost to follow-up.

No details on possible competing interests reported

HIGH (all outcomes)

Reason: lack of details on randomization, concealment of allocation, blinding, high number lost to follow-up, patients with dislocation of epidural catheter within 2 days postoperatively were excluded from study (n=7), no study protocol or registration available, no report of possible conflicts of interest

Mustola (2011)

Randomization method not reported

Reason: no details provided; “Of these, 21 were randomly allocated to a group that received thoracic epidural analgesia combined with cryoanalgesia (EC)

and 21 to a group that received the epidural without cryoanalgesia (E).”

Probably no

Reason: no details provided

Probably no

Reason: No details reported about blinding procedures; care providers and data analysts not mentioned to be blinded.

Patients and assumably assessors:

“The investigating anesthetists

(S.M., J.L.) and patients were blinded to whether cryoanalgesia

was performed.”

Probably yes

Reason: in both groups, 1 patient (4.8%) was excluded due to incomplete data. It is expected that this would not influence the results.

Probably no

Reason: no protocol or study registration reported. Relevant outcome measures reported with plausible length of follow-up.

No information

Reason:

No details on funding and possible competing interests reported.

Some concerns

(all outcomes)

Reason: lack of details on randomization, concealment of allocation, blinding, no study protocol or registration available, no report of possible conflicts of interest

Sepsas (2013)

Randomization method not reported

Reason:

“After informed consent, patients were randomly assigned to one

of two groups, using the method of closed envelopes.”

Probably no

Reason: no details provided:

Probably no

Reason: Study is called ‘double-blind’ but no details reported about blinding procedures; patients, care providers and data analysts not mentioned to be blinded.

Assessor:

“Postoperative monitoring and data recording were performed by a blinded researcher”

Definitely yes

Reason: no lost to follow-up; for both groups, n=25 included in analysis and reported up to 60 days of follow-up.

Probably no

Reason: no protocol or study registration reported. Relevant outcome measures reported with plausible length of follow-up.

No information

Reason:

No details on funding.

Some concerns

(all outcomes)

Reason: lack of details on randomization, concealment of allocation, blinding, no study protocol or registration available.