114
Injury (1991) 22, (2), 114-116
Printed in Great Britain
lntrapleural bupivacaine analgesia in chest trauma: a randomized double-blind controlled trial J. D. Knottenbelt’, M. F. James2 and M. Bloomfield ‘Trauma Unit, Groote Schuur Hospital, Cape Town, South Africa and 2Departments University of Cape Town, South Africa
A randomized double-blind study was undertaken to investigate whether 0.25 per cent bupivacaine administered infrapleurally is an effective and reasonably safe method of obtaining analgesia in patients with thoracic injuries. A total of 12Opafienfs complaining of pain uffer chest injury were entered info the trial. All had fhoracosfomy lubes already in situ. Of 60 patients who were given the test dose of bupivacaine infrapleurally, 3 i obtained satisfactory pain relief for 2 or more hours compared with only 9 of 60 patients who received saline P -C 0.000001). Duration of analgesia in the test group (mean 3.9 h) wus significantly longer than that in the control group (mean 0.9 h) P < 0.005). tie were nu important side-effects attributable to the bupivacaine administered. We conclude that intrapleural bupivacaine is an effective and reasonably safe method for obtaining analgesia in patients who have chest druin tubes inserted.
of Anaesthesia
and Surgery,
Strom&rag et al. (1988) in a study on the relief of postcholecystectomy pain, and also by Seltzer et al. (1987). The only reported adverse effects (a seizure) was with a dose three times that used in this trial. Although Rosenberg et al. (1987) had not found intrapleural bupivacaine to be useful in post-thoracotomy pain, Rocco et al. (1987) reported good results with intrapleural bupivacaine in six patients with multiple rib fractures. We had already used intrapleural bupivacaine empirically in over 100 patients without adverse effects, and were of the opinion that the benefits justified a prospective double-blind randomized study to determine its efficacy in patients with chest- trauma.
Methods Introduction Chest pain following thoracic injury interferes with the normal healing process by inhibiting chest wall movement, expectoration and coughing. In our Unit it is the policy for all patients with chest injury to take part in vigorous physiotherapy so fhat lung expansion and drainage of pleural air and fluid is obtained in minimum time. This can only be achieved if analgesia is adequate. Traditional methods of obtaining analgesia have the drawbacks outlined in Table 1. The safety of intrapleural bupivacaine and the dynamics of its distribution have recently been investigated by
Table I. Disadvantage of traditional analgesia in chest trauma
methods
of obtaining
Method
Disadvantage
1. Rib blocks
Multiple punctures Moderate expertise required Not suitable for posterior fractures Painful Time consuming Expertise required Hypotension Respiratory depression Cough reflex suppressed Often not adequate Bronchospasm gastrointestinal bleeding Anti-platelet activity
2. Thoracic epidural analgesia 3. Opiates 4. Paracetamol 5. Non-steroidal anti inflammatories
0 1991 Butteworth-Heinemann 0020-1383/91/020114-03
Ltd
A series of 120 patients admitted to the trauma unit for blunt or penetrating chest injury requiring analgesia, who already had an intercostal drain in place, were included in the trial. Informed patient consent was obtained in each case and the trial was approved by the Ethics and Research Committee of the Faculty of Medicine of the University of Cape Town. Patients were not considered for inclusion if unconscious, if known to be hypersensitive to bupivacaine or other local anaesthetic, if there were known or suspected intrapleural adhesions, cardiac injury or disease, if suffering from epilepsy or if both sides of the chest were involved in the injury. Test doses were made up by the hospital pharmacy as 10 ml of 0.5 per cent bupivacaine, with 10ml of normal saline in the control vials. Randomization was achieved using random number tables, and the sequence codes were withheld from the investigators until completion of the trial. Postural drainage of appreciable blood present in the pleura was attempted before test doses were instilled, but an
Table II. Prince Henry score for pain Score No pain on coughing Pain on coughing, not on deep breathing Pain on deep breathing but not at rest Pain at rest, slight Pain at rest, severe
Injuryz the British Journal of Accident Surgery (1991) Vol. 22/No.
2
115
Table III. Average visual analogue scores/ranks Control
Bupivacaine
0 min 5 min 15 min
Score
Rank
Score
Rank
Sig
7.01 4.65 3.77
59 50.8 48.5
7.25 6.27 5.83
62 70.2 72.5
NS Pi 0.02 P< 0.0001
Average Prince Henry scores/ranks Control
Bupivacaine
0 min 5 min 15min
Score
Rank
Score
Rank
Sig
3.25 2.05 1.57
61.7 50.4 46.6
3.22 2.82 2.63
59.3 70.6 74.4
NS P< 0.05 P< 0.0001
Sig = significance. NS = not significant.
Table V. Mean duration of pain relief
Table IV. Adequacy of pain relief
Controls Bupivacaine P< 0.000001
Failure (Up to 2 h)
Adequate (2 h or more)
51 23
379
Controls Bupivacaine P> 0.005
0.9 h (sem 0.3) 3.5 h (sem 0.44)
Results empty pleural cavity was not a condition of inclusion. The contents of one 10 ml vial was diluted with 10 ml saline and instilled into the intercostal tube with the patient lying on the contralateral side. The patient was then turned supine and the drainage tube elevated so that the instilled fluid was retained for 5 min. Thereafter the patient was mobilized in the usual way consistent with his other injuries. Subjective pain level was recorded on a visual analogue scale (VAS) of O-10 cm, before and at 5 min and 15 min after the test dose. Objective pain level was scored by the attending doctor using the Prince Henry Scale (PH) (Table U). If the patient did not experience adequate pain relief within 15 min of administration of the test dose (reduction of less than 2 cm on the VAS, or PH score above 2), the trial ended for that patient who was thereafter treated with an analgesic of the attending doctor‘s choice. This was usually one or more of paracetamol and codeine (tablets), indomethacin suppository, morphine (parenteral), local anaesthetic intercostal block, or an intrapleural dose of known bupivaCaine. If pain relief was obtained by 15 min, the VAS and PH were re-assessed at 30 min, 1, 2, 4 and 8 h. The tie when the patient required other analgesia was noted and the duration of action calculated. If pain relief lasted less than 2 h, the case was deemed a ‘failure’. VAS and PH score changes with time were assessed against test substance and statistically compared for significance using the Mann-Whitney U test. The mean duration of analgesia for each group was calculated and tested for significance using Student’s t test. Adverse side-effects were looked for routinely by electrocardiogram, pulse and blood pressure monitoring for the first 30 min after instillation.
There were 116 male and 4 female subjects aged between 15 and 75 years (mean 28.1 years, SD 11.1) entered in the trial. Of these, 6 patients had blunt injury with fractured ribs and 114 had penetrating chest wounds; 24 patients had pneumothorax, 18 had haemothorax and 78 had haemopneumothorax. Results for pain scores are given in 7iibks III-V. Fortyseven patients in the active group obtained good relief initially, but eight of these needed analgesic supplementation before 2 h. Of the 15 controls who initially had good relief, six required supplementation before 2 h. Starting VAS and PH scores for the test and control groups were not significantly different, while the ranks of both scores at 5 and 15 min were significantly in favour of the bupivacaine group. Of the 74 early ‘failures’ (who failed to obtain 2 h or more of pain relief), 21 were treated with further doses of bupivacaine intrapleurally as the choice of the attending doctor. Of the 21.14 were in the saline group and seven in the bupivacaine group. Subsequent adequate relief for 2 h or more was obtained in 20 of these patients - the nonresponder had received an initial saline test dose. The only side-effects noted were aggravation of pain in one patient in the test group, and vasovagal collapse in one of the patients receiving the control solution.
Discussion It is our policy to employ vigorous physiotherapy to encourage lung expansion and prevent atelectasis in patients requiring chest drainage for haemo- or pneumothorax. To achieve patient cooperation it is essential to provide adequate pain relief without undue sedation. This study is the first controlled trial, to our knowledge, testing the
116
Knottenbelt et al.: Intrapleural bupivacaine analgesia in chest trauma
efficacy of intrapleural bupivacaine in the treatment of post-traumatic chest pain. Our results show that 20 ml intrapleural bupivacaine 0.25 per cent is highly effective in providing analgesia in most patients with haemo- or pneumothorax. The good response to a further dose of bupivacaine in those test group failures treated in this way suggests that supplementation of the dose may be needed in some patients - studies on blood levels suggest that up to 40 ml of 0.25per cent bupivacaine (100 mg) should not raise plasma bupivacaine levels above 1.8 pg/ml, well below the 4.0pg/ml safety limit (Seltzer et al., 1987; Strom&rag et al., 1988). Apart from inadequate dose, a possible cause for failure of the technique could be dilution with residual blood in the pleural cavity. The beneficial effect seen with intrapleural saline in some of the controls was probably due to the placebo effect, but might also have been the resulf of lubrication of the pleural membranes or dilution of local inflammatory substances. All of our patients had a chest drain inserted for other indications, and this was probably a contributory factor in the production of pain. At present we do not insert chest catheters for the sole purpose of providing analgesia. Our results, together with those of Rocco et al. (1987), suggest that this may well be justified in certain patients who have severe pain following chest trauma.
Acknowledgements We are grateful to the Chief Medical Groote Schuur Hospital for permission
Superintendent of to publish patient
data, to Sapharmed for financial support and supply of the and to the pharmaceutical staff of active bupivacaine, Groote Schuur Hospital for preparation of the test doses.
References RoccoA.,
Reiestad F., Gudman J. et al. (1987)Intrapleural administration of local anesthetics for pain relief in patients with multiple rib fractures. Reg. Anaesth. 12, 10. Rosenberg H., Scheinin B. M.-A., Lepantalo M. J. A. et al. (1987) Continuous intrapleural infusion of bupivacaine for analgesia after thoracotomy. Anesfhesiology 67, 81 I. Seltzer J. L., Larijani G. E.. Goldberg M. E. et al. (1987) Intrapleural bupivacaine - a kinetic and dynamic evaluation. Anesfhesiologg
67,798. Strom&rag K. E., Reiestad F., Holmqvist E. L. 0. et al. (1988) Intrapleural administration of 0.25%, 0.375%, and 0.5%bupivaCaine with epinephrine after cholecystectomy. Anesth. Analg.
67,430.
Paper accepted 11 June 1990.
Requests for reprints should be addressed to: Dr J. D. Knottenbelt, Trauma Unit, Groote Schuur Hospital, Observatory 7925,Cape Town, South Africa.
Injury (1991) 22, (2), 114-116
Printed in Great Britain
lntrapleural bupivacaine analgesia in chest trauma: a randomized double-blind controlled trial J. D. Knottenbelt’, M. F. James2 and M. Bloomfield ‘Trauma Unit, Groote Schuur Hospital, Cape Town, South Africa and 2Departments University of Cape Town, South Africa
A randomized double-blind study was undertaken to investigate whether 0.25 per cent bupivacaine administered infrapleurally is an effective and reasonably safe method of obtaining analgesia in patients with thoracic injuries. A total of 12Opafienfs complaining of pain uffer chest injury were entered info the trial. All had fhoracosfomy lubes already in situ. Of 60 patients who were given the test dose of bupivacaine infrapleurally, 3 i obtained satisfactory pain relief for 2 or more hours compared with only 9 of 60 patients who received saline P -C 0.000001). Duration of analgesia in the test group (mean 3.9 h) wus significantly longer than that in the control group (mean 0.9 h) P < 0.005). tie were nu important side-effects attributable to the bupivacaine administered. We conclude that intrapleural bupivacaine is an effective and reasonably safe method for obtaining analgesia in patients who have chest druin tubes inserted.
of Anaesthesia
and Surgery,
Strom&rag et al. (1988) in a study on the relief of postcholecystectomy pain, and also by Seltzer et al. (1987). The only reported adverse effects (a seizure) was with a dose three times that used in this trial. Although Rosenberg et al. (1987) had not found intrapleural bupivacaine to be useful in post-thoracotomy pain, Rocco et al. (1987) reported good results with intrapleural bupivacaine in six patients with multiple rib fractures. We had already used intrapleural bupivacaine empirically in over 100 patients without adverse effects, and were of the opinion that the benefits justified a prospective double-blind randomized study to determine its efficacy in patients with chest- trauma.
Methods Introduction Chest pain following thoracic injury interferes with the normal healing process by inhibiting chest wall movement, expectoration and coughing. In our Unit it is the policy for all patients with chest injury to take part in vigorous physiotherapy so fhat lung expansion and drainage of pleural air and fluid is obtained in minimum time. This can only be achieved if analgesia is adequate. Traditional methods of obtaining analgesia have the drawbacks outlined in Table 1. The safety of intrapleural bupivacaine and the dynamics of its distribution have recently been investigated by
Table I. Disadvantage of traditional analgesia in chest trauma
methods
of obtaining
Method
Disadvantage
1. Rib blocks
Multiple punctures Moderate expertise required Not suitable for posterior fractures Painful Time consuming Expertise required Hypotension Respiratory depression Cough reflex suppressed Often not adequate Bronchospasm gastrointestinal bleeding Anti-platelet activity
2. Thoracic epidural analgesia 3. Opiates 4. Paracetamol 5. Non-steroidal anti inflammatories
0 1991 Butteworth-Heinemann 0020-1383/91/020114-03
Ltd
A series of 120 patients admitted to the trauma unit for blunt or penetrating chest injury requiring analgesia, who already had an intercostal drain in place, were included in the trial. Informed patient consent was obtained in each case and the trial was approved by the Ethics and Research Committee of the Faculty of Medicine of the University of Cape Town. Patients were not considered for inclusion if unconscious, if known to be hypersensitive to bupivacaine or other local anaesthetic, if there were known or suspected intrapleural adhesions, cardiac injury or disease, if suffering from epilepsy or if both sides of the chest were involved in the injury. Test doses were made up by the hospital pharmacy as 10 ml of 0.5 per cent bupivacaine, with 10ml of normal saline in the control vials. Randomization was achieved using random number tables, and the sequence codes were withheld from the investigators until completion of the trial. Postural drainage of appreciable blood present in the pleura was attempted before test doses were instilled, but an
Table II. Prince Henry score for pain Score No pain on coughing Pain on coughing, not on deep breathing Pain on deep breathing but not at rest Pain at rest, slight Pain at rest, severe
Injuryz the British Journal of Accident Surgery (1991) Vol. 22/No.
2
115
Table III. Average visual analogue scores/ranks Control
Bupivacaine
0 min 5 min 15 min
Score
Rank
Score
Rank
Sig
7.01 4.65 3.77
59 50.8 48.5
7.25 6.27 5.83
62 70.2 72.5
NS Pi 0.02 P< 0.0001
Average Prince Henry scores/ranks Control
Bupivacaine
0 min 5 min 15min
Score
Rank
Score
Rank
Sig
3.25 2.05 1.57
61.7 50.4 46.6
3.22 2.82 2.63
59.3 70.6 74.4
NS P< 0.05 P< 0.0001
Sig = significance. NS = not significant.
Table V. Mean duration of pain relief
Table IV. Adequacy of pain relief
Controls Bupivacaine P< 0.000001
Failure (Up to 2 h)
Adequate (2 h or more)
51 23
379
Controls Bupivacaine P> 0.005
0.9 h (sem 0.3) 3.5 h (sem 0.44)
Results empty pleural cavity was not a condition of inclusion. The contents of one 10 ml vial was diluted with 10 ml saline and instilled into the intercostal tube with the patient lying on the contralateral side. The patient was then turned supine and the drainage tube elevated so that the instilled fluid was retained for 5 min. Thereafter the patient was mobilized in the usual way consistent with his other injuries. Subjective pain level was recorded on a visual analogue scale (VAS) of O-10 cm, before and at 5 min and 15 min after the test dose. Objective pain level was scored by the attending doctor using the Prince Henry Scale (PH) (Table U). If the patient did not experience adequate pain relief within 15 min of administration of the test dose (reduction of less than 2 cm on the VAS, or PH score above 2), the trial ended for that patient who was thereafter treated with an analgesic of the attending doctor‘s choice. This was usually one or more of paracetamol and codeine (tablets), indomethacin suppository, morphine (parenteral), local anaesthetic intercostal block, or an intrapleural dose of known bupivaCaine. If pain relief was obtained by 15 min, the VAS and PH were re-assessed at 30 min, 1, 2, 4 and 8 h. The tie when the patient required other analgesia was noted and the duration of action calculated. If pain relief lasted less than 2 h, the case was deemed a ‘failure’. VAS and PH score changes with time were assessed against test substance and statistically compared for significance using the Mann-Whitney U test. The mean duration of analgesia for each group was calculated and tested for significance using Student’s t test. Adverse side-effects were looked for routinely by electrocardiogram, pulse and blood pressure monitoring for the first 30 min after instillation.
There were 116 male and 4 female subjects aged between 15 and 75 years (mean 28.1 years, SD 11.1) entered in the trial. Of these, 6 patients had blunt injury with fractured ribs and 114 had penetrating chest wounds; 24 patients had pneumothorax, 18 had haemothorax and 78 had haemopneumothorax. Results for pain scores are given in 7iibks III-V. Fortyseven patients in the active group obtained good relief initially, but eight of these needed analgesic supplementation before 2 h. Of the 15 controls who initially had good relief, six required supplementation before 2 h. Starting VAS and PH scores for the test and control groups were not significantly different, while the ranks of both scores at 5 and 15 min were significantly in favour of the bupivacaine group. Of the 74 early ‘failures’ (who failed to obtain 2 h or more of pain relief), 21 were treated with further doses of bupivacaine intrapleurally as the choice of the attending doctor. Of the 21.14 were in the saline group and seven in the bupivacaine group. Subsequent adequate relief for 2 h or more was obtained in 20 of these patients - the nonresponder had received an initial saline test dose. The only side-effects noted were aggravation of pain in one patient in the test group, and vasovagal collapse in one of the patients receiving the control solution.
Discussion It is our policy to employ vigorous physiotherapy to encourage lung expansion and prevent atelectasis in patients requiring chest drainage for haemo- or pneumothorax. To achieve patient cooperation it is essential to provide adequate pain relief without undue sedation. This study is the first controlled trial, to our knowledge, testing the
116
Knottenbelt et al.: Intrapleural bupivacaine analgesia in chest trauma
efficacy of intrapleural bupivacaine in the treatment of post-traumatic chest pain. Our results show that 20 ml intrapleural bupivacaine 0.25 per cent is highly effective in providing analgesia in most patients with haemo- or pneumothorax. The good response to a further dose of bupivacaine in those test group failures treated in this way suggests that supplementation of the dose may be needed in some patients - studies on blood levels suggest that up to 40 ml of 0.25per cent bupivacaine (100 mg) should not raise plasma bupivacaine levels above 1.8 pg/ml, well below the 4.0pg/ml safety limit (Seltzer et al., 1987; Strom&rag et al., 1988). Apart from inadequate dose, a possible cause for failure of the technique could be dilution with residual blood in the pleural cavity. The beneficial effect seen with intrapleural saline in some of the controls was probably due to the placebo effect, but might also have been the resulf of lubrication of the pleural membranes or dilution of local inflammatory substances. All of our patients had a chest drain inserted for other indications, and this was probably a contributory factor in the production of pain. At present we do not insert chest catheters for the sole purpose of providing analgesia. Our results, together with those of Rocco et al. (1987), suggest that this may well be justified in certain patients who have severe pain following chest trauma.
Acknowledgements We are grateful to the Chief Medical Groote Schuur Hospital for permission
Superintendent of to publish patient
data, to Sapharmed for financial support and supply of the and to the pharmaceutical staff of active bupivacaine, Groote Schuur Hospital for preparation of the test doses.
References RoccoA.,
Reiestad F., Gudman J. et al. (1987)Intrapleural administration of local anesthetics for pain relief in patients with multiple rib fractures. Reg. Anaesth. 12, 10. Rosenberg H., Scheinin B. M.-A., Lepantalo M. J. A. et al. (1987) Continuous intrapleural infusion of bupivacaine for analgesia after thoracotomy. Anesfhesiology 67, 81 I. Seltzer J. L., Larijani G. E.. Goldberg M. E. et al. (1987) Intrapleural bupivacaine - a kinetic and dynamic evaluation. Anesfhesiologg
67,798. Strom&rag K. E., Reiestad F., Holmqvist E. L. 0. et al. (1988) Intrapleural administration of 0.25%, 0.375%, and 0.5%bupivaCaine with epinephrine after cholecystectomy. Anesth. Analg.
67,430.
Paper accepted 11 June 1990.
Requests for reprints should be addressed to: Dr J. D. Knottenbelt, Trauma Unit, Groote Schuur Hospital, Observatory 7925,Cape Town, South Africa.
