- Email: [email protected]
Blast injuries to the lungs: clinical presentation, management and course
INJURY Volume 8/Number
1
The British Journal of Accident Surgery
Blast injuries to the lungs: clinical presentation, management and course N. G. Caseby
and M. F. Porter
Senior Reaistrar in Anaesthetics* Birmingha”m Accident Hospital
and Consultant Surgeon,
Summary
CASE
Five patients with blast injuries to the lungs after bomb explosions are reported. In each patient radiological changes were apparent on the initial chest film taken within 4 hours of the explosions. Arterial hypoxaemia was also present. Four patients were actively treated with continuous positive-pressure ventilation, which was adjudged effective therapy. Two patients died, one owing to bilateral pneumothorax which occurred during anaesthesia, and the other owing to overwhelming infection. Hypoxaemia persisted for 4 months in one of the survivors. Lung function tests which were performed on the same patient 10 months after the blast injuries, however, were normal.
Case 1
INTRODUCTION IN A recent article it was suggested
that the clinical and radiological features of blast injuries to the lungs, and the changes which occur in the arterial blood gases, do not appear for 24-48 hours after an explosion (Gray and Coppel, 1975). This has not been our experience with 5 patients who sustained lung damage when 2 bombs went off almost simultaneously in 2 public houses in Birmingham on 21 November, 1974. These patients were admitted to hospital within 30 minutes of the incidents and were all conscious on arrival. Each patient had multiple lacerations and splinter wounds (which contained much debris), plus superficial burns, in addition to the injuries specified in the individual case histories. * Present address: Department University of Birmingham.
of
Anaesthetics,
REPORTS
This 34-year-old male was in severe respiratory distress and was deeply cyanosed. His systolic blood pressure (BP) was 100 mm Hg. Air entry to his right lung was diminished and diffuse crepitations were heard in this lung. He had many soft-tissue wounds of his right side including one particularly large wound of the chest (but without pleural perforation). A drain was promptly inserted in the right pleural cavity. Ten minutes later, because of continuing respiratory distress, the patient was intubated and positive-pressure ventilation (PPV) was started. Blood was aspirated from the tracheobronchial tree. A chest X-ray 60 minutes after the explosion revealed a large opacity in the right lower zone and a fracture of the right sixth rib (Fig. 1). Peritoneal lavage 30 minutes later indicated intra-abdominal bleeding. A large rent on the superior surface of the liver was found at laparotomy and this was sutured. Primary excision of his other wounds, without closure, was also carried out. By the end of the operation the patient had received 22 units of blood and 2 1 of Hartmann’s solution. After the operation PPV was reinstituted with an East-Freeman ventilator. A positive end-expiratory pressure (PEEP) of 10 cm H,O was added. For most of his ventilator therapy a fractional inspired oxygen concentration (Fro,) of 0.3 resulted in an arterial oxygen tension (Pao,) of between 9.3 and 10.65 kPa (70 and 80 mm Hg). The maximal inflation pressure reached on delivering a tidal volume (VT) of 1000 ml was 35 cm H,O. A left pneumothorax developed on the fourth day after the explosion and a drain was accordingly inserted in the pleural cavity. By the ninth day considerable clinical and radiological improvement in the pulmonary condition had occurred and, therefore, weaning from the ventilator was started. 1
2
Injury: the British Journal
Fig. I. Case 1. Initial chest X-ray film showing right lower zone, maximal peripherally.
multiple
foreign
Fig. 2. Case 1. Chest X-ray On the eleventh day, however, the patient became drowsy and weak. Full ventilator control was reestablished and the FIO, was increased to 0.4. A right pneumothorax (Fig. 2) which arose on day 12 was treated. Tracheostomy was performed on day 13. A chest X-ray on the fourteenth day showed the lung fields to be clear except for a little residual shadowing
bodies
of Accident
in chest
Surgery
Vol. ~/NO. 1
wall and dense shadowing
film on day 12 showing
in
right pneumothorax.
at the right base (Fig. 3). The patient remained crititally ill, however, and this was attributed to septicaemia. His general condition got worse and on day 17 he developed hypotension which was resistant to therapy. He died on the eighteenth day. The diagnosis of septicaemia was confirmed at autopsy. Both liver and spleen were grossly enlarged.
Caseby and Porter: Blast Injuries to Lungs
Fig. 3. Case 1. Chest X-ray film on day 14 showing re-expansion shadowing.
of right lung with a little residual basal
Fig. 4. Case 2. Initial (70 minutes) chest X-ray film showing bilateral and symmetrical shadowing of ‘bats-wing’ distribution. A swab was taken from splenic tissue and it grew all the organisms which had been cultured during life from his soft-tissue wounds. Foci of bronchopneumonit consolidation were found in both lungs and some subpleural bullae were noted in the apex of the right upper lobe.
Case 2 This 20-year-old female complained of severe retrosternal chest pain and difficulty in breathing. Her BP was 90/60 mm Hg, pulse rate 140 per minute, and respiratory rate 30 per minute. Her main external injury was a mangled right foot. A chest X-ray 70
4
Injury: the British Journal
minutes after the explosion showed pulmonary infiltrates in the mid and lower zones of both lungs (Fig. 4). She began to cough up small amounts of fresh blood. Diffuse crepitations were heard throughout both lungs. Six hours after sustaining the blast injuries her Paoz was 8.51 kPa (64 mm Hg) and Pacoz 5.39 kPa (40.5 mm Hg). General anaesthesia was induced 30 minutes later for primary excision of wounds, including mid-tarsal amputation of the
of Accident
Surgery
Vol. ~/NO. 1
right foot. without closure. Blood was aspirated from the tracheobronchial tree several times during anaesthesia. By the end of the operation I3 units of blood had been given. After the operation PPV therapy was started with an East-Radcliffe ventilator and a PEEP of IOcm H,O was added. PEEP was reduced to 5 cm H,O on day 5 and to zero on day 6. During the ventilator therapy a FIO* of 0.3 resulted in a PaoP of about 13.3
Fig. 5. Case 2. Chest
X-ray
film on day 4 showing
Fig. 6. Case 2. Chest
X-ray
film on day
11showing
slight clearing.
further
clearing
Caseby
and Porter:
Blast Injuries
to Lungs
kPa (100 mm Hg) and the peak inflation pressure reached on delivering a VT of 800 ml was 30 cm H,O. Slight radiological clearing of the lung fields was noted on day 4 (Fig. 5). On the seventh day PPV was discontinued and the patient was extubated. Oxygen was given with a Ventimask for several days after the ventilator therapy because she still had arterial hypoxaemia. Her Pao, values while breathing air were 9.84: 8.25 and 11.17 kPa (74: 62 and 84 mm Hg) on the days 8, 10 and 13 respectively. A chest film on day 11 showed the lung fields to be almost clear although some infiltrates were still present in both cases (Fig. 6). By the time of her discharge from hospital 27 days after the explosion, complete radiological resolution had occurred but her Pao, was still low at 11.17 kPa (84 mm Hg).
5
added, his systolic BP fell from 130 to 90 mm Hg and, therefore, PEEP was removed. A further 3 units of blood were transfused (making a total of 11 units) and reapplication of PEEP (IO cm H,O) 2 hours later had no detrimental effect. PEEP was reduced to 5 cm H,O on the third day after the explosion and to zero on the fourth. During the ventilator therapy a FIO~ of 0.25-0.3 resulted in a Pao, of about 13.3 kPa (100 mm Hg), and the maximal inflation pressure reached on delivering a VT of 1100 ml was 25 cm H,O. Some clearing of the lung fields was noted on day 3. On the fourth day PPV was discontinued and he was extubated. The pulmonary condition was clinically much improved but his Paoz while breathing air was only 7.45 kPa (56 mm Hg). Oxygen was, therefore, given with a Ventimask for several more
Fig. 7. Case 3. Initial chest X-ray filnn show Gng ill-defined shadowing spreading outwards from the Iright hilum and to a lesser degree from the left. Case 3 This 31-year-old male had external injuries which were confined to his legs. He complained, however, of retrosternal chest pain and difficulty in breathing. His BP was 120/80 mm Hg. uulse rate 64 per minute and respiratory rate 28 p&-minute. A chest X-ray 90 minutes after the explosion showed pulmonary infiltrates spreading throughout the right lung from the hilum, and some patchy infiltration in the left lung (Fig. 7). A fracture of the right tibia was reduced and primary wound excision performed under general anaesthesia 4 hours after sustaining the blast injuries. During anaesthesia blood was aspirated from the tracheobronchial tree and lung crepitations were heard. After the operation PPV therapy was started with a Cape ventilator. When a PEEP of 10 cm HZ0 was
days. Complete radiological resolution was noted on day 6. On day 14 his Pao, was 10.91 kPa (82 mm Hg). Case 4 This 28-year-old male complained of severe retrosternal chest pain and dyspnoea. His respiratory rate was 36 per minute and he was slightly cyanosed. His BP was 100/60 mm Hg and pulse rate 120 per minute. Diffuse crepitations and expiratory rhonchi were heard in both lungs. His main external injury was a 50 per cent partial-thickness burn which involved mainly the anterior aspect of his whole body. A chest X-ray taken within 2 hours of the explosion revealed a well-defined round opacity of 8 cm diameter in the left mid zone, with less clearly defined opacities fanning out above and below it (Fig. 8). His Paoz
Injury: the British Journal
Fig. 8. Case 4. Initial c:hest X-ray film showing a circumscribed clearly defined shadowing above and below it. _
opacity
of Accident
adjacent
Surgery
Vol. ~/NO. 1
to the left hilum
with
less
14 hours after injury were 5-05, 8.78 and 7.45 kPa (38. 66 and 56 mm HE) resnectivelv while breathing air. The corresponding-iaco, results were 4.06, 3.88 and 4.72 kPa (30.5, 29.2 and 35.5 mm
small lacerations
Hg). Primary excision of wounds and dressing of burns were performed 20 hours after the explosion. During general anaesthesia blood was aspirated from the tracheobronchial tree. PPV therapy was then started and a PEEP of 10 cm H,O was added. A FIO, of 0.35-0.4 was required to maintain a PaoB above 9-3 kPa (70 mm Hg), and a
This 21-year-old male complained of chest pain, both retrosternal and in the left costophrenic region, and difficulty in taking a deep breath. His BP was 15OjlOO mm Hg, pulse rate 130 per minute, and respiratory rate 24 per minute. Air entry to the left lung was diminished and crepitations were heard in this lung. His main injury was a 20 per cent partial-thickness burn of his legs and back. A chest X-ray at 4 hours showed an opacity in the left lower zone with less and spreading outwards well-defined opacities upwards from the left hilum, plus fractures of the seventh and eighth ribs on the left (Fig. 9). Six hours after injury his Pao, was 8.25 kPa (62 mm Hg) and Pace., 4.59 kPa (34.5 mm He). Primary excision of wounds and dressjng of burnswere carried out under general anaesthesia 12 hours later. This patient was treated with bed rest and oxygen.
values at 6, 10 and
maximal inflation pressure of 40 cm H,O was reached on delivering a V, of 1100 ml. Bronchospasm, which was present before PPV was begun, continued. Subcutaneous surgical emphysema and a right pneumothorax developed on the second day and, therefore, a drain was inserted in the right pleural cavity. Six hours later collapse of the right lower and middle lobes occurred. Thick mucous plugs were removed at bronchoscopy. On the third day the patient became pyrexial. His urinary sodium excretion fell to Immol/ litre on days 4 and 5 and, since this was associated with a normal blood volume (5.2 I): septicaemia was suspected (Jackson, 1975). On the fifth day the right chest drain was removed. As no improvement in the patient’s clinical condition had occurred by day 6, despite slight radiological clearing of the lung fields, a tracheostomy was performed. Anaesthesia was provided with 50 per cent nitrous oxide in oxygen and PPV was continued. Towards the end of the operation the patient had a cardiac arrest. Resuscitation was unsuccessful. Autopsy revealed bilateral pneumothorax. Areas of contusion and foci of bronchopneumonic consolidation were present in both lungs. There were several
of the visceral
pleura
near the right
hilum. Case 5
For 3 days he coughed up blood-stained bronchial secretions. Complete radiological resolution was noted on day 8. By day 11 he was asymptomatic and his Pao,
was 10.64 kPa (80 mm Hg).
DISCUSSION injuries have been classified as primary, secondary or tertiary (White et al., 1964; de Candole, 1967; Hamit, 1973). Primary blast injuries are those which are inflicted by the pressure wave itself; secondary blast injuries are those which result from the patient being struck Blast
7
Caseby and Porter: Blast Injuries to Lungs
Fig. 9. Case 5. Initial chest X-ray film showing shadowing fanning outwards and upwards from the left hilum.
by flying debris or falling masonry; and tertiary blast injuries are those which are sustained when a patient is violently thrown. Primary blast injury to the lungs results in tearing of the tissues of the lungs and causes most damage to the alveolar parenchyma. The pressure wave may also produce tears in the visceral pleura, as well as bullae and alveolar-venous fistulae formation (Zuckerman, 1940; Carhon et al., 1945; White et al., 1964; de Candole, 1967; Hamit, 1973). Secondary or tertiary blast injuries to the chest may cause lung damage from either a penetrating injury of the chest or, more frequently, a crush injury. The commonest complication of a crush injury to the chest is pulmonary contusion, and this lesion is characterized by destruction of alveolar structure (Sykes et al., 1969; Fraser and Pare, 1970). Thus, in blast injuries to the lungs, whatever the cause of the lung damage, the predominant pulmonary lesion consists of alveolar parenchymal damage with exudation of oedema fluid and blood into the interstitial space and alveoli. The main pathological difference between lung damage due to primary blast injury and that due to crush injury is probably the increased likelihood of bullae, alveolar-venous fistulae and pleural tears in the former. The pulmonary autopsy findings after death by blast are well documented (Hadfield et al., 1940; Wilson and Tunbridge, 1943; Waterworth and Carr, 1975) and correspond
at the left base and less well-defined shadowing
closely with experimental 1940; Rossle, 1950).
CLINICAL
pathology
(Zuckerman,
PRESENTATION
A diagnosis of blast injuries to the lungs, based on clinical and radiological findings, was made in each patient in this report within a few hours of the explosions. The clinical features which our patients exhibited are similar to those described by others (O’Reilly and Gloyne, 1941; Hadfield and Christie, 1941; Huller and Bazini, 1970). Each patient was X-rayed 14 hours after the incident and in every case gross pulmonary changes were present on the initial chest film. The radiological findings were those of pulmonary contusion and oedema. Hirsch and Bazini (1969) also found changes on the first chest films of 1 I patients less than 9 hours after an underwater explosion. Arterial blood gas analysis was carried out on 3 of our young patients (Cases 2, 4 and 5) 6 hours after the blast injuries. Each patient had arterial hypoxaemia. No arterial sample was taken from Case 1 (who was clinically in respiratory failure) before PPV therapy was started, nor from Case 3.
MANAGEMENT The intra-alveolar oedema and haemorrhage plus alveolar atelectasis (from plugging of the bronchioles with blood clot) that occur in blast
8
injuries to the lungs probably result in a decrease in functional residual capacity (FRC), a lowering of the ventilation : perfusion ratio (3 : d), and an increase in right-to-left shunting of blood through non-ventilated alveoli. A minor degree of lung damage may cause slight impairment of gas exchange with resultant mild arterial hypoxaemia, whereas a more extensive pulmonary lesion may produce severe impairment of gas exchange and result in acute respiratory failure, The main objective of treatment of blast injuries to the lungs is to restore arterial blood gases to near normal until the lungs have recovered. This may be achieved with controlled oxygen therapy delivered by a face mask, but, where severe lung damage has been sustained, PPV with variable oxygen concentrations is often indicated. Some authors, however, believe that PPV is contra-indicated in the treatment of primary blast injury to the lungs because of the possibility of introducing bubbles of gas into the pulmonary veins through alveolar-venous fistulae (Hamit, 1973; Weiler-Ravel] et al., 1975). The existence of these fistulae has been demonstrated microscopically (White et al., 1964) and there is no doubt that air embolism can occur at the time of an explosion and is a major cause of sudden death (Carlton et al., 1945; Benzinger, 1950). What is less clear is how long the patient is at risk from air embolism after the initial impact of the blast wave, and how long local reflex mechanisms such as pulmonary venoconstriction take to produce spontaneous closure of the fistulae. During PPV a substantial part of the pressure required to inflate the lungs is dissipated in overcoming resistance in the bronchial tree, and the changes in pressure in the alveoli are smaller than those in the main airways. The pressure changes in alveoli which have been damaged by blast are probably even smaller owing to the alveoli being filled with blood and oedema fluid. It is unlikely, therefore, that PPV will increase the risk of air embolism in the management of blast injuries to the lungs, but the patient should be examined regularly for evidence of embolic phenomena. The development of focal neurological signs or ECG signs of transient myocardial ischaemia must arouse suspicion of air emboli in the cerebral or coronary arteries respectively. Ophthalmoscopy may reveal air bubbles in the retinal arteries. Another objection to the use of PPV in blast injuries to the lungs is that patients are liable to develop pneumothorax, but this is readily treated and therefore should not contra-indicate the use of PPV. Hyperbaric therapy has been
Injury: the British Journal
of Accident
Surgery
Vol. ~/NO.
1
suggested as an alternative to PPV (WeilerRavel] et al., 1975), but hyperbaric facilities are not readily or universally available and there is no convincing evidence that this treatment is superior to PPV. Weiler-Ravel1 et al. have also discussed the use of a membrane oxygenator. Whatever form of therapy is used, reduction of the patient’s oxygen consumption by simple measures such as bed rest and sedation or cooling to normal if pyrexial may help torelieve thedegree of arterial hypoxaemia. positive-pressure ventilation Continuous (CPPV) has been advocated (Gray and Coppel, 1975) and used successfully (McCaughey et al., 1973) in the treatment of respiratory failure due to blast injuries to the lungs. This is interesting because theoretically CPPV is likely to increase further the risk of air embolism and pneumothorax owing to the continuously raised intraalveolar pressure produced by the technique. The magnitude of these dangers is unknown, but the question arises as to whether the advantages of using a PEEP in the treatment of blast injuries to the lungs outweigh the theoretical disadvantages. The main benefit from PEEP is the improvement in arterial oxygenation which has followed its use in some patients with acute respiratory failure. This increase in PaoB may be life-saving, and in many patients it permits the Fro, to be reduced and may diminish the problem of pulmonary oxygen toxicity. The mechanism by which PEEP improves Pao, is probably by increasing the FRC above the critical closing volume and thereby decreasing the shunt of blood flowing across non-ventilated alveoli (Ashbaugh et al., 1969; Powers et al., 1973; Suter et al., 1975). PEEP is likely to benefit patients with a decreased FRC and a consequent need to clear fluid-filled alveoli or open up collapsed alveoli (Suter et al., 1975), and may, therefore, be advantageous in blast injuries to the lungs. The application of a PEEP, however, may be detrimental in that it may reduce cardiac output (0.r) (Powers et al., 1973) or overdistend alveoli and so decrease lung compliance and eventually cause rupture and pneumothorax (Kumar et al., 1970). A small fall in eT may result in the total quantity of oxygen delivered to the tissues being reduced, even though PaoB is increased as a result of PEEP (Suter et al., 1975). A larger fall in 8.i causes a greater reduction in systemic oxygen transport because Paon may be decreased by reduced saturation of venous blood (Kelman et al., 1967). This adverse effect by of PEEP on or, however, can be corrected augmenting the blood volume (Qvist et al., 1975).
Caseby and Porter:
Blast Injuries
to Lungs
PPV was employed in the treatment of 4 of our patients. Case 1, who was in severe respiratory distress, was artificially ventilated as soon as it was ascertained that his clinical features were not due to some correctable cause such as haemopneumothorax. The other patients (Cases 2, 3 and 4) were ventilated following their respective operations and after assessment of the severity of their pulmonary lesions, based on radiological and clinical findings. Although only one of these patients was in hypoxaemic respiratory failure (PaoZ less than 8 kPa or 60 mm Hg) before operation, it was decided to institute early elective PPV rather than await further deterioration in arterial blood gases. During ventilator therapy the FIO, was adjusted to achieve a Paoz of between 9.3 and 13.3 kPa (70 and 100 mm Hg), and the patient’s Pacoz was maintained between 4 and 5.3 kPa (30 and 40 mm Hg). The inspired inflation pressure reached on delivering a large l’/T was not found to be unduly high The FIO~ requirements of our patients throughout the main phase of their pulmonary injuries ranged from 0.25 to 0.4, and this contrasted with the high oxygen concentrations needed by 2 cases reported by McCaughey et al. (1973). Only 1 of our patients (Case I) received more than 40 per cent oxygen, and that was given during his terminal illness, more than 2 weeks after the blast injuries. A PEEP of 1Ocm Hz0 was applied to each patient a few hours after ventilation began and was continued until clinical and radiological improvement of the pulmonary lesion had occurred, when it was reduced in stages to zero. After the addition of PEEP each patient was monitored to detect any adverse effects, such as haemodynamic deterioration, decrease in effective compliance (Shires et al., 1973), or signs of air embolism. No attempt was made to assess optimal or ‘best’ PEEP (Suter et al., 1975) for individual patients and little objective evidence of the benefit of PEEP is available. It was felt, however, that CPPV was advantageous because first, adequate arterial oxygenation was achieved with relatively small FIO, values (not more than 0.4); secondly, the patients tolerated the ventilator well and required minimal sedation (papaveretum); and thirdly, satisfactory radiological resolution occurred. Some of the complications associated with the use of CPPV in blast injuries to the lungs were met: the initial application of PEEP to Case 3 resulted in hypotension due to hypovolaemia; pneumothorax occurred twice in both Cases 1 and 4, and the development of bilateral pneumothorax in Case 4 during general
9
anaesthesia proved to be a fatal event; bronchial secretions became infected in Cases 1 and 4; and surgical emphysema and lobar atelectasis occurred in Case 4. No symptoms or signs of air embolism were detected in any patient during ventilation. Because of the beneficial effects of CPPV and the lack of evidence of air embolism, we conclude, from our limited number of cases, that CPPV is definitely indicated in the treatment of blast injuries to the lungs. We stress that CPPV should be used with caution and that doctors should be alert to the possible development of pneumothorax or air embolism. A case has been made for prophylactic bilateral pleural drainage (McCaughey et al., 1973), but as the use of chest tubes is not without its own complications, we feel that careful monitoring is preferable. If a pneumothorax does occur, however, we recommend that both pleural cavities be drained and that the drainage be maintained throughout the period of ventilation. Monitoring is particularly essential during operations under general anaesthesia, where access to the patient is often considerably restricted. One patient in this report (Case 5) was treated with oxygen given by a face mask. Although he was hypoxaemic when breathing air (Pao, 8.25 kPa or 62 mm Hg), he was managed conservatively for two reasons: first, his radiological changes were considered less severe than those of the other patients; and secondly, blood was not aspirated from his tracheobronchial tree at operation. His progress was satisfactory and his respiratory state did not deteriorate. There is no specific treatment for blast injuries to the lungs. Removal of blood and secretions from the tracheobronchial tree by suction, together with chest physiotherapy to re-expand atelectatic areas, may increase the p : Q ratio and thereby improve oxygenation. Corticosteroid and diuretic therapy have been advocated by Gray and Coppel (1975). Diuretics were not necessary in the management of our patients. We gave 4-6 doses of methylprednisolone (SoluMedrone) 250 mg, however, to 3 patients (Cases I, 2 and 3) throughout the first 2 days of their treatment, but it is difficult to assess whether the steroids had any beneficial effect on the pulmonary lesions. Infection is, unfortunately, a major hazard in patients with multiple secondary or tertiary blast injuries or burns. For example, many organisms were cultured from the softtissue wounds of our patients; Cases 1 and 4 were each suspected of having septicaemia, and Case 1 undoubtedly died from it. As repeated doses of
10
steroid can alter immune mechanisms and lower resistance to infection, they should perhaps be avoided in blast injuries until sufficient evidence is available to support their use. Patients with blast injuries to the lungs may be hypovolaemic owing to blood or plasma loss from other associated injuries. Hypovolaemia may result in a fall in eT and decrease the systemic venous saturation, which will aggravate any hypoxia due to lung damage (Kelman et al., 1937; Pontoppidan et al., 1970). In addition, hypovolaemic patients are more susceptible to abrupt falls in QT with ventilator therapy (Maloney et al., 1951; Hubay et al., 1954). Immediate blood or fluid replacement and the restoration of intravascular volume is, therefore, mandatory. An assessment of the circulating volume can be obtained, where heart function is normal, by monitoring the central venous pressure (CVP) response to the rapid administration of 200 ml fluid. A small transient rise in CVP suggests continuing hypovolaemia; a small but more sustained rise, normovolaemia; and a larger persistent rise, hypervolaemia (Sykes, 1963). It has been postulated, however, that pulmonary vascular resistance may be raised in patients with primary blast injury to the lungs and that the work load of the right ventricle may be increased (Weiler-Ravel1 et al., 1975). If this supposition is correct then a CVP which is higher than normal may be required to achieve maximum performance of the right heart. This optimal CVP can be determined by correlating changes of CVP with the clinical signs of 07: a fluid load which produces a small rise in CVP together with an increase in Q1 is acceptable, but one which produces a rise in CVP without any change or a decrease in e T is indicative of impairment of the pump action of the heart (Cohn, 1967). Excessive quantities of crystalloid fluid should be avoided in patients with blast injuries to the lungs. These solutions rapidly equilibrate with interstitial fluid and will accumulate preferentially in the lung tissues where there is increased capillary permeability. This increase in lung extravascular water may produce an increase in pulmonary shunt which will further impair pulmonary gas exchange (Obdrzalek et al., 1975). Replacement of circulating volume is best achieved by giving a fluid similar in composition to that lost.
COURSE The course of blast injuries to the lungs consists of a stage of pulmonary bleeding followed by a
Injury:
the British
Journal of Accident Surgery Vol. ~/NO. 1
stage of coagulation, absorption and repair (Wilson, 1943). In our patients the stage of bleeding lasted about 24 hours, during which time fresh blood was aspirated in diminishing quantities from those who were intubated. The stage of repair of the pulmonary lesion was characterized by the disappearance of lung crepitations. a reduction in ventilator inflation pressure, improved gas exchange as judged by decreasing oxygen requirements to achieve adequate PaoP levels, and radiological resolution. About 2 weeks after the blast injuries, pulmonary gas exchange was much improved. This can be illustrated by blood gas analysis which was carried out on the 3 surviving patients after treatment had been discontinued. The Paoz values of Cases 2,3 and 5 on days 13, 14 and 11 after the explosion were 11.17, IO.91 and IO.64 kPa (84, 82 and 80 mm Hg) respectively. The patients were breathing air spontaneously and their Pacol values were normal. This degree of arterial hypoxaemia indicated that only partial repair of the pulmonary lesion had taken place. Further arterial blood sampling was, therefore, carried out on one patient (Case 2) to find out how long it would take for complete repair to occur. Her Paon 27 days after the blast injuries was still low at I I.17 kPa (84 mm Hg), and it took slightly over 4 months for the Pao, to reach a normal value (13.3 kPa or 100 mm Hg). The chief complaint in the 3 survivors was retrosternal chest pain or discomfort, which lasted for about 10 days in Cases 3 and 5, and for more than 1 month in Case 2. Five months after the explosion Case 3 still had chest pain and dyspnoea on exercise. The radiological course of blast injuries to the lungs is consistent except where complications such as pneumothorax, atelectasis or infection arise. Progression in the severity of pulmonary contusion is unusual after 6 hours so that the appearance of a more extensive lesion after 48 hours suggests a superimposed disease process (Stevens and Templeton, 1965; Fraser and Pare, 1970). Resolution may be rapid where the blast injuries have resulted in exudation of oedema fluid, but it takes longer where haemorrhage is present (Williams and Stembridge, 1964). The intitial clearing of the lung fields in our patients occurred in 24 days. McGrigor and Samuel (1945) report that much resolution takes place in about 3 days, and in 10 days or so the lung fields appear normal on the radiograph. This time course is in keeping with our findings in Cases 3 and 5, but the infiltrates in a third patient who survived (Case 2) took longer to clear.
11
Caseby and Porter: Blast Injuries to Lungs Residual lung damage in the form of emphysema is said to occur in animals who have survived an exposure to blast (de Candole, 1967). The possibility of lasting pulmonary injury was investigated in Case 2 by lung function tests, including residual volume and transfer factor for carbon monoxide, 10 months after the explosion. These tests were all normal. It was concluded that there is no evidence of permanent lung damage of any consequence in this patient. Our experience in treating 5 patients with blast injuries to the lungs suggests that the prognosis is fair, providing that PPV therapy can be instituted if required until adequate parenchymal repair has taken place, and that the development of any pulmonary complication is quickly dealt with. However, in patients who have received additional soft-tissue wounds from secondary and tertiary blast injuries or extensive burns, infection and septicaemia become a problem. Acknowledgements Our thanks are due to our consultant colleagues at the Birmingham Accident Hospital for permission to study cases under their care, to Dr S. Sevitt and Dr H. Thompson for post-mortem details, to Dr P. Jacobs for radiological interpretation, to Dr J. F. Riordan for performing lung function studies in Case 2 and to Mr N. R. Gill for photographic reproduction. REFERENCES Ashbaugh D. C., Petty T. L., Bigelow D. B. et al. (1969) Continuous positive-pressure breathing (CPPB) in adult respiratory distress syndrome. J. Thorac. Cardiovasc. Sup. 57, 3 1. Benzinger T. (1950) Physiological effects of blast in air and water. In: German Aviation Medicine World War II, vol. 2. Washington, US Government Printing Office, p. 1225. de Candole C. A. (1967) Blast injury. Can. Med. Assoc. J. 96, 207.
Carlton L. M. jun., Rasmussen R. A. and Adams W. E. (1945) Blast injury of lung: possible explanation of mechanism in fatal cases--experimental study. Surgery 17, 786. Cohn J. N. (1967) Central venous pressure as a guide to volume expansion. Ann. Intern. Med. 66, 1283. Fraser R. G. and Pare J. A. P. (1970) Diagnosis of Diseases of the Chest, vol. 2. Philadelphia, Saunders, p. 1069. Gray R. C. and Coppel D. L. (1975) Surgery of violence. III. Intensive care of patients with bomb blast and gunshot injuries. Br. Med. J. 1,502. Hadfield G. and Christie R. V. (1941) A case of pulmonary concussion (‘blast’) due to high explosive Br. Med. J. 1, 77.
Hadfield G., Ross J. M., Swain R. H. A. et al. (1940) Blast from high explosive. Preliminary report on ten fatal cases. Lancet 2, 478. Hamit H. F. (1973) Primary blast injuries. Ind. Med. Surg. 42, No. 3, 14.
Hirsch M. and Bazini J. (1969) Blast injury to thechest. Clin. Radiol. 20, 362.
Hubav C. A.. Waltz R. C.. Brecher G. A. et al. (1954) Circulatory dynamics ‘of venous return during positive-negative pressure respiration. Anesthesiology 15, 445.
Huller T. and Bazini J. (1970) Blast injuries of the chest and abdomen. Arch. Surg. 100,24. Jackson D. M. (1975) The sick cell syndrome in burns. In: Vrabec R., Konickova Z. and Moserova J. (ed.) Basic Problems in Burns. Prague, Avicenum, p. 174. Kelman G. R., Nunn J. F., Prys-Roberts C. et al. (1967) The influence of cardiac output on arterial oxygenation: a theoretical study. Br. J. Anaesth. 39, 450.
Kumar A., Falke K. J., Gefhn B. et al. (1970) Continuous positive-pressure ventilation in acute respiratory failure. Effects on hemodynamics and lung function. N. Engl. J. Med. 283, 1430. McCaughey W., Coppel D. L. and Dundee J. W. (1973) Blast injuries to the lungs. A report of two cases. Anaesthesia 28, 2. McGrigor D. B. and Samuel E. (1945) The radiology of war injuries. Part Ill. War wounds of the chest. Br. J. Radiol. 18,133. Maloney J. V. jun., Affeldt J. E., Sarnoff S. J. et al. (1951) Electrophrenic respiration. IX. Comparison of effects of positive-pressure breathing and electrophrenic respiration on the circulation during haemorrhagic shock and barbiturate poisoning. Surg. Gynecol.
Obstet.
92, 672.
Obdrzalek J., Kay J. C. and Noble W. H. (1975) The effects of continuous positive-pressure ventilation on pulmonary oedema, gas exchange and lung mechanics. Can. Anaesth. Sot. J. 22, 399. O’Reilly J. N. and Gloyne S. R. (1941) Blast injury of the lungs. Lancer 2, 423. Pontoppidan H., Laver M. B. and Geffin B. (1970) Acute respiratory failure in the surgical patient. In: Welch C. (ed.) Advances in Surgery, vol. 4. Chicago, Year Book, p. 163. Powers S. R. jun., Mannal R., Neclerio M. et al. (1973) Physiologic consequences of positive endexpiratory pressure (PEEP) ventilation. Ann. Surg. 178, 265.
Qvist J., Pontoppidan H., Wilson R. S. et al. (1975) Hemodynamic responses to mechanical ventilation with PEEP: the effect of hypervolemia. Anesthesioloav 42. 45.
RiissG R.’ (1950) Pathology German
Aviation
Medicine
of blast World
effects.
War II, vol.
In: 2.
Washington, US Government Printing Office, p. 1260. Shires G. T., Carrico C. J. and Canizaro P. C. (1973) Shock : Major Problems in Clinical Surgery, vol. 13. Philadelphia, Saunders, p. 88.
12
Injury:the British Journal
Stevens E. and Templeton A. W. (1965) Traumatic non-penetrating lung contusion. Radiology 85, 247. Suter P. M., Fairley H. B. and lsenberg M. D. (1975) Optimum end-expiratory airway pressure in patients with acute pulmonary failure. N. Engl. J. Med. 292, 284. Sykes M. K. (1963) Venous pressure as a clinical indication of adequacy of transfusion. Ann. R. Co//. Surg. Engl. 33, 185. Sykes M. K., McNicol M. W. and Campbell E. J. M. (1969) Respirafory Failure. Oxford, Blackwell Scientific, p. 215. Waterworth T. A. and Carr M. J. T. (1975) An analysis of the post-mortem findings in the 21 victims of the Birmingham pub bombings. lnury 7, 89.
Requests for reprints should be addressed to: Dr N. Queen Elizabeth Hospital, Birmingham, 815 2TH.
G.
Caseby,
of Accident
Surgery
Vol. ~/NO. 1
Weiler-Ravel1 D., Adatto R. and Borman J. B. (1975) Blast injury of the chest. A review of the problem and its treatment. Isr. J. Med. Sri. 11, 268. White C. S., Bowen 1. G. and Richmond D. R. (1964) A comparative analysis of some of the immediate environmental effects at Hiroshima and Nagasaki. Health Phys. 10, 89. Williams J. R. and Stembridge V. A. (1964) Pulmonary contusion secondary to non-penetrating chest trauma. Am. J. Roentgenol. Radium Ther. Nucl. Med. 91, 284. Wilson J. V. (1943) The pathology of closed injuries of the chest. Br. Med. J. 1, 470. Wilson J. V. and Tunbridge R. E. (1943) Pathological findings in a series of blast injuries. Lancer 1, 257. Zuckerman S. (1940) Experimental study of blast injuries to the lungs. Lancet 2, 219.
FFA
RCS,
Department
of Anaesthetics,
University
of‘ Rirmingham,
1
The British Journal of Accident Surgery
Blast injuries to the lungs: clinical presentation, management and course N. G. Caseby
and M. F. Porter
Senior Reaistrar in Anaesthetics* Birmingha”m Accident Hospital
and Consultant Surgeon,
Summary
CASE
Five patients with blast injuries to the lungs after bomb explosions are reported. In each patient radiological changes were apparent on the initial chest film taken within 4 hours of the explosions. Arterial hypoxaemia was also present. Four patients were actively treated with continuous positive-pressure ventilation, which was adjudged effective therapy. Two patients died, one owing to bilateral pneumothorax which occurred during anaesthesia, and the other owing to overwhelming infection. Hypoxaemia persisted for 4 months in one of the survivors. Lung function tests which were performed on the same patient 10 months after the blast injuries, however, were normal.
Case 1
INTRODUCTION IN A recent article it was suggested
that the clinical and radiological features of blast injuries to the lungs, and the changes which occur in the arterial blood gases, do not appear for 24-48 hours after an explosion (Gray and Coppel, 1975). This has not been our experience with 5 patients who sustained lung damage when 2 bombs went off almost simultaneously in 2 public houses in Birmingham on 21 November, 1974. These patients were admitted to hospital within 30 minutes of the incidents and were all conscious on arrival. Each patient had multiple lacerations and splinter wounds (which contained much debris), plus superficial burns, in addition to the injuries specified in the individual case histories. * Present address: Department University of Birmingham.
of
Anaesthetics,
REPORTS
This 34-year-old male was in severe respiratory distress and was deeply cyanosed. His systolic blood pressure (BP) was 100 mm Hg. Air entry to his right lung was diminished and diffuse crepitations were heard in this lung. He had many soft-tissue wounds of his right side including one particularly large wound of the chest (but without pleural perforation). A drain was promptly inserted in the right pleural cavity. Ten minutes later, because of continuing respiratory distress, the patient was intubated and positive-pressure ventilation (PPV) was started. Blood was aspirated from the tracheobronchial tree. A chest X-ray 60 minutes after the explosion revealed a large opacity in the right lower zone and a fracture of the right sixth rib (Fig. 1). Peritoneal lavage 30 minutes later indicated intra-abdominal bleeding. A large rent on the superior surface of the liver was found at laparotomy and this was sutured. Primary excision of his other wounds, without closure, was also carried out. By the end of the operation the patient had received 22 units of blood and 2 1 of Hartmann’s solution. After the operation PPV was reinstituted with an East-Freeman ventilator. A positive end-expiratory pressure (PEEP) of 10 cm H,O was added. For most of his ventilator therapy a fractional inspired oxygen concentration (Fro,) of 0.3 resulted in an arterial oxygen tension (Pao,) of between 9.3 and 10.65 kPa (70 and 80 mm Hg). The maximal inflation pressure reached on delivering a tidal volume (VT) of 1000 ml was 35 cm H,O. A left pneumothorax developed on the fourth day after the explosion and a drain was accordingly inserted in the pleural cavity. By the ninth day considerable clinical and radiological improvement in the pulmonary condition had occurred and, therefore, weaning from the ventilator was started. 1
2
Injury: the British Journal
Fig. I. Case 1. Initial chest X-ray film showing right lower zone, maximal peripherally.
multiple
foreign
Fig. 2. Case 1. Chest X-ray On the eleventh day, however, the patient became drowsy and weak. Full ventilator control was reestablished and the FIO, was increased to 0.4. A right pneumothorax (Fig. 2) which arose on day 12 was treated. Tracheostomy was performed on day 13. A chest X-ray on the fourteenth day showed the lung fields to be clear except for a little residual shadowing
bodies
of Accident
in chest
Surgery
Vol. ~/NO. 1
wall and dense shadowing
film on day 12 showing
in
right pneumothorax.
at the right base (Fig. 3). The patient remained crititally ill, however, and this was attributed to septicaemia. His general condition got worse and on day 17 he developed hypotension which was resistant to therapy. He died on the eighteenth day. The diagnosis of septicaemia was confirmed at autopsy. Both liver and spleen were grossly enlarged.
Caseby and Porter: Blast Injuries to Lungs
Fig. 3. Case 1. Chest X-ray film on day 14 showing re-expansion shadowing.
of right lung with a little residual basal
Fig. 4. Case 2. Initial (70 minutes) chest X-ray film showing bilateral and symmetrical shadowing of ‘bats-wing’ distribution. A swab was taken from splenic tissue and it grew all the organisms which had been cultured during life from his soft-tissue wounds. Foci of bronchopneumonit consolidation were found in both lungs and some subpleural bullae were noted in the apex of the right upper lobe.
Case 2 This 20-year-old female complained of severe retrosternal chest pain and difficulty in breathing. Her BP was 90/60 mm Hg, pulse rate 140 per minute, and respiratory rate 30 per minute. Her main external injury was a mangled right foot. A chest X-ray 70
4
Injury: the British Journal
minutes after the explosion showed pulmonary infiltrates in the mid and lower zones of both lungs (Fig. 4). She began to cough up small amounts of fresh blood. Diffuse crepitations were heard throughout both lungs. Six hours after sustaining the blast injuries her Paoz was 8.51 kPa (64 mm Hg) and Pacoz 5.39 kPa (40.5 mm Hg). General anaesthesia was induced 30 minutes later for primary excision of wounds, including mid-tarsal amputation of the
of Accident
Surgery
Vol. ~/NO. 1
right foot. without closure. Blood was aspirated from the tracheobronchial tree several times during anaesthesia. By the end of the operation I3 units of blood had been given. After the operation PPV therapy was started with an East-Radcliffe ventilator and a PEEP of IOcm H,O was added. PEEP was reduced to 5 cm H,O on day 5 and to zero on day 6. During the ventilator therapy a FIO* of 0.3 resulted in a PaoP of about 13.3
Fig. 5. Case 2. Chest
X-ray
film on day 4 showing
Fig. 6. Case 2. Chest
X-ray
film on day
11showing
slight clearing.
further
clearing
Caseby
and Porter:
Blast Injuries
to Lungs
kPa (100 mm Hg) and the peak inflation pressure reached on delivering a VT of 800 ml was 30 cm H,O. Slight radiological clearing of the lung fields was noted on day 4 (Fig. 5). On the seventh day PPV was discontinued and the patient was extubated. Oxygen was given with a Ventimask for several days after the ventilator therapy because she still had arterial hypoxaemia. Her Pao, values while breathing air were 9.84: 8.25 and 11.17 kPa (74: 62 and 84 mm Hg) on the days 8, 10 and 13 respectively. A chest film on day 11 showed the lung fields to be almost clear although some infiltrates were still present in both cases (Fig. 6). By the time of her discharge from hospital 27 days after the explosion, complete radiological resolution had occurred but her Pao, was still low at 11.17 kPa (84 mm Hg).
5
added, his systolic BP fell from 130 to 90 mm Hg and, therefore, PEEP was removed. A further 3 units of blood were transfused (making a total of 11 units) and reapplication of PEEP (IO cm H,O) 2 hours later had no detrimental effect. PEEP was reduced to 5 cm H,O on the third day after the explosion and to zero on the fourth. During the ventilator therapy a FIO~ of 0.25-0.3 resulted in a Pao, of about 13.3 kPa (100 mm Hg), and the maximal inflation pressure reached on delivering a VT of 1100 ml was 25 cm H,O. Some clearing of the lung fields was noted on day 3. On the fourth day PPV was discontinued and he was extubated. The pulmonary condition was clinically much improved but his Paoz while breathing air was only 7.45 kPa (56 mm Hg). Oxygen was, therefore, given with a Ventimask for several more
Fig. 7. Case 3. Initial chest X-ray filnn show Gng ill-defined shadowing spreading outwards from the Iright hilum and to a lesser degree from the left. Case 3 This 31-year-old male had external injuries which were confined to his legs. He complained, however, of retrosternal chest pain and difficulty in breathing. His BP was 120/80 mm Hg. uulse rate 64 per minute and respiratory rate 28 p&-minute. A chest X-ray 90 minutes after the explosion showed pulmonary infiltrates spreading throughout the right lung from the hilum, and some patchy infiltration in the left lung (Fig. 7). A fracture of the right tibia was reduced and primary wound excision performed under general anaesthesia 4 hours after sustaining the blast injuries. During anaesthesia blood was aspirated from the tracheobronchial tree and lung crepitations were heard. After the operation PPV therapy was started with a Cape ventilator. When a PEEP of 10 cm HZ0 was
days. Complete radiological resolution was noted on day 6. On day 14 his Pao, was 10.91 kPa (82 mm Hg). Case 4 This 28-year-old male complained of severe retrosternal chest pain and dyspnoea. His respiratory rate was 36 per minute and he was slightly cyanosed. His BP was 100/60 mm Hg and pulse rate 120 per minute. Diffuse crepitations and expiratory rhonchi were heard in both lungs. His main external injury was a 50 per cent partial-thickness burn which involved mainly the anterior aspect of his whole body. A chest X-ray taken within 2 hours of the explosion revealed a well-defined round opacity of 8 cm diameter in the left mid zone, with less clearly defined opacities fanning out above and below it (Fig. 8). His Paoz
Injury: the British Journal
Fig. 8. Case 4. Initial c:hest X-ray film showing a circumscribed clearly defined shadowing above and below it. _
opacity
of Accident
adjacent
Surgery
Vol. ~/NO. 1
to the left hilum
with
less
14 hours after injury were 5-05, 8.78 and 7.45 kPa (38. 66 and 56 mm HE) resnectivelv while breathing air. The corresponding-iaco, results were 4.06, 3.88 and 4.72 kPa (30.5, 29.2 and 35.5 mm
small lacerations
Hg). Primary excision of wounds and dressing of burns were performed 20 hours after the explosion. During general anaesthesia blood was aspirated from the tracheobronchial tree. PPV therapy was then started and a PEEP of 10 cm H,O was added. A FIO, of 0.35-0.4 was required to maintain a PaoB above 9-3 kPa (70 mm Hg), and a
This 21-year-old male complained of chest pain, both retrosternal and in the left costophrenic region, and difficulty in taking a deep breath. His BP was 15OjlOO mm Hg, pulse rate 130 per minute, and respiratory rate 24 per minute. Air entry to the left lung was diminished and crepitations were heard in this lung. His main injury was a 20 per cent partial-thickness burn of his legs and back. A chest X-ray at 4 hours showed an opacity in the left lower zone with less and spreading outwards well-defined opacities upwards from the left hilum, plus fractures of the seventh and eighth ribs on the left (Fig. 9). Six hours after injury his Pao, was 8.25 kPa (62 mm Hg) and Pace., 4.59 kPa (34.5 mm He). Primary excision of wounds and dressjng of burnswere carried out under general anaesthesia 12 hours later. This patient was treated with bed rest and oxygen.
values at 6, 10 and
maximal inflation pressure of 40 cm H,O was reached on delivering a V, of 1100 ml. Bronchospasm, which was present before PPV was begun, continued. Subcutaneous surgical emphysema and a right pneumothorax developed on the second day and, therefore, a drain was inserted in the right pleural cavity. Six hours later collapse of the right lower and middle lobes occurred. Thick mucous plugs were removed at bronchoscopy. On the third day the patient became pyrexial. His urinary sodium excretion fell to Immol/ litre on days 4 and 5 and, since this was associated with a normal blood volume (5.2 I): septicaemia was suspected (Jackson, 1975). On the fifth day the right chest drain was removed. As no improvement in the patient’s clinical condition had occurred by day 6, despite slight radiological clearing of the lung fields, a tracheostomy was performed. Anaesthesia was provided with 50 per cent nitrous oxide in oxygen and PPV was continued. Towards the end of the operation the patient had a cardiac arrest. Resuscitation was unsuccessful. Autopsy revealed bilateral pneumothorax. Areas of contusion and foci of bronchopneumonic consolidation were present in both lungs. There were several
of the visceral
pleura
near the right
hilum. Case 5
For 3 days he coughed up blood-stained bronchial secretions. Complete radiological resolution was noted on day 8. By day 11 he was asymptomatic and his Pao,
was 10.64 kPa (80 mm Hg).
DISCUSSION injuries have been classified as primary, secondary or tertiary (White et al., 1964; de Candole, 1967; Hamit, 1973). Primary blast injuries are those which are inflicted by the pressure wave itself; secondary blast injuries are those which result from the patient being struck Blast
7
Caseby and Porter: Blast Injuries to Lungs
Fig. 9. Case 5. Initial chest X-ray film showing shadowing fanning outwards and upwards from the left hilum.
by flying debris or falling masonry; and tertiary blast injuries are those which are sustained when a patient is violently thrown. Primary blast injury to the lungs results in tearing of the tissues of the lungs and causes most damage to the alveolar parenchyma. The pressure wave may also produce tears in the visceral pleura, as well as bullae and alveolar-venous fistulae formation (Zuckerman, 1940; Carhon et al., 1945; White et al., 1964; de Candole, 1967; Hamit, 1973). Secondary or tertiary blast injuries to the chest may cause lung damage from either a penetrating injury of the chest or, more frequently, a crush injury. The commonest complication of a crush injury to the chest is pulmonary contusion, and this lesion is characterized by destruction of alveolar structure (Sykes et al., 1969; Fraser and Pare, 1970). Thus, in blast injuries to the lungs, whatever the cause of the lung damage, the predominant pulmonary lesion consists of alveolar parenchymal damage with exudation of oedema fluid and blood into the interstitial space and alveoli. The main pathological difference between lung damage due to primary blast injury and that due to crush injury is probably the increased likelihood of bullae, alveolar-venous fistulae and pleural tears in the former. The pulmonary autopsy findings after death by blast are well documented (Hadfield et al., 1940; Wilson and Tunbridge, 1943; Waterworth and Carr, 1975) and correspond
at the left base and less well-defined shadowing
closely with experimental 1940; Rossle, 1950).
CLINICAL
pathology
(Zuckerman,
PRESENTATION
A diagnosis of blast injuries to the lungs, based on clinical and radiological findings, was made in each patient in this report within a few hours of the explosions. The clinical features which our patients exhibited are similar to those described by others (O’Reilly and Gloyne, 1941; Hadfield and Christie, 1941; Huller and Bazini, 1970). Each patient was X-rayed 14 hours after the incident and in every case gross pulmonary changes were present on the initial chest film. The radiological findings were those of pulmonary contusion and oedema. Hirsch and Bazini (1969) also found changes on the first chest films of 1 I patients less than 9 hours after an underwater explosion. Arterial blood gas analysis was carried out on 3 of our young patients (Cases 2, 4 and 5) 6 hours after the blast injuries. Each patient had arterial hypoxaemia. No arterial sample was taken from Case 1 (who was clinically in respiratory failure) before PPV therapy was started, nor from Case 3.
MANAGEMENT The intra-alveolar oedema and haemorrhage plus alveolar atelectasis (from plugging of the bronchioles with blood clot) that occur in blast
8
injuries to the lungs probably result in a decrease in functional residual capacity (FRC), a lowering of the ventilation : perfusion ratio (3 : d), and an increase in right-to-left shunting of blood through non-ventilated alveoli. A minor degree of lung damage may cause slight impairment of gas exchange with resultant mild arterial hypoxaemia, whereas a more extensive pulmonary lesion may produce severe impairment of gas exchange and result in acute respiratory failure, The main objective of treatment of blast injuries to the lungs is to restore arterial blood gases to near normal until the lungs have recovered. This may be achieved with controlled oxygen therapy delivered by a face mask, but, where severe lung damage has been sustained, PPV with variable oxygen concentrations is often indicated. Some authors, however, believe that PPV is contra-indicated in the treatment of primary blast injury to the lungs because of the possibility of introducing bubbles of gas into the pulmonary veins through alveolar-venous fistulae (Hamit, 1973; Weiler-Ravel] et al., 1975). The existence of these fistulae has been demonstrated microscopically (White et al., 1964) and there is no doubt that air embolism can occur at the time of an explosion and is a major cause of sudden death (Carlton et al., 1945; Benzinger, 1950). What is less clear is how long the patient is at risk from air embolism after the initial impact of the blast wave, and how long local reflex mechanisms such as pulmonary venoconstriction take to produce spontaneous closure of the fistulae. During PPV a substantial part of the pressure required to inflate the lungs is dissipated in overcoming resistance in the bronchial tree, and the changes in pressure in the alveoli are smaller than those in the main airways. The pressure changes in alveoli which have been damaged by blast are probably even smaller owing to the alveoli being filled with blood and oedema fluid. It is unlikely, therefore, that PPV will increase the risk of air embolism in the management of blast injuries to the lungs, but the patient should be examined regularly for evidence of embolic phenomena. The development of focal neurological signs or ECG signs of transient myocardial ischaemia must arouse suspicion of air emboli in the cerebral or coronary arteries respectively. Ophthalmoscopy may reveal air bubbles in the retinal arteries. Another objection to the use of PPV in blast injuries to the lungs is that patients are liable to develop pneumothorax, but this is readily treated and therefore should not contra-indicate the use of PPV. Hyperbaric therapy has been
Injury: the British Journal
of Accident
Surgery
Vol. ~/NO.
1
suggested as an alternative to PPV (WeilerRavel] et al., 1975), but hyperbaric facilities are not readily or universally available and there is no convincing evidence that this treatment is superior to PPV. Weiler-Ravel1 et al. have also discussed the use of a membrane oxygenator. Whatever form of therapy is used, reduction of the patient’s oxygen consumption by simple measures such as bed rest and sedation or cooling to normal if pyrexial may help torelieve thedegree of arterial hypoxaemia. positive-pressure ventilation Continuous (CPPV) has been advocated (Gray and Coppel, 1975) and used successfully (McCaughey et al., 1973) in the treatment of respiratory failure due to blast injuries to the lungs. This is interesting because theoretically CPPV is likely to increase further the risk of air embolism and pneumothorax owing to the continuously raised intraalveolar pressure produced by the technique. The magnitude of these dangers is unknown, but the question arises as to whether the advantages of using a PEEP in the treatment of blast injuries to the lungs outweigh the theoretical disadvantages. The main benefit from PEEP is the improvement in arterial oxygenation which has followed its use in some patients with acute respiratory failure. This increase in PaoB may be life-saving, and in many patients it permits the Fro, to be reduced and may diminish the problem of pulmonary oxygen toxicity. The mechanism by which PEEP improves Pao, is probably by increasing the FRC above the critical closing volume and thereby decreasing the shunt of blood flowing across non-ventilated alveoli (Ashbaugh et al., 1969; Powers et al., 1973; Suter et al., 1975). PEEP is likely to benefit patients with a decreased FRC and a consequent need to clear fluid-filled alveoli or open up collapsed alveoli (Suter et al., 1975), and may, therefore, be advantageous in blast injuries to the lungs. The application of a PEEP, however, may be detrimental in that it may reduce cardiac output (0.r) (Powers et al., 1973) or overdistend alveoli and so decrease lung compliance and eventually cause rupture and pneumothorax (Kumar et al., 1970). A small fall in eT may result in the total quantity of oxygen delivered to the tissues being reduced, even though PaoB is increased as a result of PEEP (Suter et al., 1975). A larger fall in 8.i causes a greater reduction in systemic oxygen transport because Paon may be decreased by reduced saturation of venous blood (Kelman et al., 1967). This adverse effect by of PEEP on or, however, can be corrected augmenting the blood volume (Qvist et al., 1975).
Caseby and Porter:
Blast Injuries
to Lungs
PPV was employed in the treatment of 4 of our patients. Case 1, who was in severe respiratory distress, was artificially ventilated as soon as it was ascertained that his clinical features were not due to some correctable cause such as haemopneumothorax. The other patients (Cases 2, 3 and 4) were ventilated following their respective operations and after assessment of the severity of their pulmonary lesions, based on radiological and clinical findings. Although only one of these patients was in hypoxaemic respiratory failure (PaoZ less than 8 kPa or 60 mm Hg) before operation, it was decided to institute early elective PPV rather than await further deterioration in arterial blood gases. During ventilator therapy the FIO, was adjusted to achieve a Paoz of between 9.3 and 13.3 kPa (70 and 100 mm Hg), and the patient’s Pacoz was maintained between 4 and 5.3 kPa (30 and 40 mm Hg). The inspired inflation pressure reached on delivering a large l’/T was not found to be unduly high The FIO~ requirements of our patients throughout the main phase of their pulmonary injuries ranged from 0.25 to 0.4, and this contrasted with the high oxygen concentrations needed by 2 cases reported by McCaughey et al. (1973). Only 1 of our patients (Case I) received more than 40 per cent oxygen, and that was given during his terminal illness, more than 2 weeks after the blast injuries. A PEEP of 1Ocm Hz0 was applied to each patient a few hours after ventilation began and was continued until clinical and radiological improvement of the pulmonary lesion had occurred, when it was reduced in stages to zero. After the addition of PEEP each patient was monitored to detect any adverse effects, such as haemodynamic deterioration, decrease in effective compliance (Shires et al., 1973), or signs of air embolism. No attempt was made to assess optimal or ‘best’ PEEP (Suter et al., 1975) for individual patients and little objective evidence of the benefit of PEEP is available. It was felt, however, that CPPV was advantageous because first, adequate arterial oxygenation was achieved with relatively small FIO, values (not more than 0.4); secondly, the patients tolerated the ventilator well and required minimal sedation (papaveretum); and thirdly, satisfactory radiological resolution occurred. Some of the complications associated with the use of CPPV in blast injuries to the lungs were met: the initial application of PEEP to Case 3 resulted in hypotension due to hypovolaemia; pneumothorax occurred twice in both Cases 1 and 4, and the development of bilateral pneumothorax in Case 4 during general
9
anaesthesia proved to be a fatal event; bronchial secretions became infected in Cases 1 and 4; and surgical emphysema and lobar atelectasis occurred in Case 4. No symptoms or signs of air embolism were detected in any patient during ventilation. Because of the beneficial effects of CPPV and the lack of evidence of air embolism, we conclude, from our limited number of cases, that CPPV is definitely indicated in the treatment of blast injuries to the lungs. We stress that CPPV should be used with caution and that doctors should be alert to the possible development of pneumothorax or air embolism. A case has been made for prophylactic bilateral pleural drainage (McCaughey et al., 1973), but as the use of chest tubes is not without its own complications, we feel that careful monitoring is preferable. If a pneumothorax does occur, however, we recommend that both pleural cavities be drained and that the drainage be maintained throughout the period of ventilation. Monitoring is particularly essential during operations under general anaesthesia, where access to the patient is often considerably restricted. One patient in this report (Case 5) was treated with oxygen given by a face mask. Although he was hypoxaemic when breathing air (Pao, 8.25 kPa or 62 mm Hg), he was managed conservatively for two reasons: first, his radiological changes were considered less severe than those of the other patients; and secondly, blood was not aspirated from his tracheobronchial tree at operation. His progress was satisfactory and his respiratory state did not deteriorate. There is no specific treatment for blast injuries to the lungs. Removal of blood and secretions from the tracheobronchial tree by suction, together with chest physiotherapy to re-expand atelectatic areas, may increase the p : Q ratio and thereby improve oxygenation. Corticosteroid and diuretic therapy have been advocated by Gray and Coppel (1975). Diuretics were not necessary in the management of our patients. We gave 4-6 doses of methylprednisolone (SoluMedrone) 250 mg, however, to 3 patients (Cases I, 2 and 3) throughout the first 2 days of their treatment, but it is difficult to assess whether the steroids had any beneficial effect on the pulmonary lesions. Infection is, unfortunately, a major hazard in patients with multiple secondary or tertiary blast injuries or burns. For example, many organisms were cultured from the softtissue wounds of our patients; Cases 1 and 4 were each suspected of having septicaemia, and Case 1 undoubtedly died from it. As repeated doses of
10
steroid can alter immune mechanisms and lower resistance to infection, they should perhaps be avoided in blast injuries until sufficient evidence is available to support their use. Patients with blast injuries to the lungs may be hypovolaemic owing to blood or plasma loss from other associated injuries. Hypovolaemia may result in a fall in eT and decrease the systemic venous saturation, which will aggravate any hypoxia due to lung damage (Kelman et al., 1937; Pontoppidan et al., 1970). In addition, hypovolaemic patients are more susceptible to abrupt falls in QT with ventilator therapy (Maloney et al., 1951; Hubay et al., 1954). Immediate blood or fluid replacement and the restoration of intravascular volume is, therefore, mandatory. An assessment of the circulating volume can be obtained, where heart function is normal, by monitoring the central venous pressure (CVP) response to the rapid administration of 200 ml fluid. A small transient rise in CVP suggests continuing hypovolaemia; a small but more sustained rise, normovolaemia; and a larger persistent rise, hypervolaemia (Sykes, 1963). It has been postulated, however, that pulmonary vascular resistance may be raised in patients with primary blast injury to the lungs and that the work load of the right ventricle may be increased (Weiler-Ravel1 et al., 1975). If this supposition is correct then a CVP which is higher than normal may be required to achieve maximum performance of the right heart. This optimal CVP can be determined by correlating changes of CVP with the clinical signs of 07: a fluid load which produces a small rise in CVP together with an increase in Q1 is acceptable, but one which produces a rise in CVP without any change or a decrease in e T is indicative of impairment of the pump action of the heart (Cohn, 1967). Excessive quantities of crystalloid fluid should be avoided in patients with blast injuries to the lungs. These solutions rapidly equilibrate with interstitial fluid and will accumulate preferentially in the lung tissues where there is increased capillary permeability. This increase in lung extravascular water may produce an increase in pulmonary shunt which will further impair pulmonary gas exchange (Obdrzalek et al., 1975). Replacement of circulating volume is best achieved by giving a fluid similar in composition to that lost.
COURSE The course of blast injuries to the lungs consists of a stage of pulmonary bleeding followed by a
Injury:
the British
Journal of Accident Surgery Vol. ~/NO. 1
stage of coagulation, absorption and repair (Wilson, 1943). In our patients the stage of bleeding lasted about 24 hours, during which time fresh blood was aspirated in diminishing quantities from those who were intubated. The stage of repair of the pulmonary lesion was characterized by the disappearance of lung crepitations. a reduction in ventilator inflation pressure, improved gas exchange as judged by decreasing oxygen requirements to achieve adequate PaoP levels, and radiological resolution. About 2 weeks after the blast injuries, pulmonary gas exchange was much improved. This can be illustrated by blood gas analysis which was carried out on the 3 surviving patients after treatment had been discontinued. The Paoz values of Cases 2,3 and 5 on days 13, 14 and 11 after the explosion were 11.17, IO.91 and IO.64 kPa (84, 82 and 80 mm Hg) respectively. The patients were breathing air spontaneously and their Pacol values were normal. This degree of arterial hypoxaemia indicated that only partial repair of the pulmonary lesion had taken place. Further arterial blood sampling was, therefore, carried out on one patient (Case 2) to find out how long it would take for complete repair to occur. Her Paon 27 days after the blast injuries was still low at I I.17 kPa (84 mm Hg), and it took slightly over 4 months for the Pao, to reach a normal value (13.3 kPa or 100 mm Hg). The chief complaint in the 3 survivors was retrosternal chest pain or discomfort, which lasted for about 10 days in Cases 3 and 5, and for more than 1 month in Case 2. Five months after the explosion Case 3 still had chest pain and dyspnoea on exercise. The radiological course of blast injuries to the lungs is consistent except where complications such as pneumothorax, atelectasis or infection arise. Progression in the severity of pulmonary contusion is unusual after 6 hours so that the appearance of a more extensive lesion after 48 hours suggests a superimposed disease process (Stevens and Templeton, 1965; Fraser and Pare, 1970). Resolution may be rapid where the blast injuries have resulted in exudation of oedema fluid, but it takes longer where haemorrhage is present (Williams and Stembridge, 1964). The intitial clearing of the lung fields in our patients occurred in 24 days. McGrigor and Samuel (1945) report that much resolution takes place in about 3 days, and in 10 days or so the lung fields appear normal on the radiograph. This time course is in keeping with our findings in Cases 3 and 5, but the infiltrates in a third patient who survived (Case 2) took longer to clear.
11
Caseby and Porter: Blast Injuries to Lungs Residual lung damage in the form of emphysema is said to occur in animals who have survived an exposure to blast (de Candole, 1967). The possibility of lasting pulmonary injury was investigated in Case 2 by lung function tests, including residual volume and transfer factor for carbon monoxide, 10 months after the explosion. These tests were all normal. It was concluded that there is no evidence of permanent lung damage of any consequence in this patient. Our experience in treating 5 patients with blast injuries to the lungs suggests that the prognosis is fair, providing that PPV therapy can be instituted if required until adequate parenchymal repair has taken place, and that the development of any pulmonary complication is quickly dealt with. However, in patients who have received additional soft-tissue wounds from secondary and tertiary blast injuries or extensive burns, infection and septicaemia become a problem. Acknowledgements Our thanks are due to our consultant colleagues at the Birmingham Accident Hospital for permission to study cases under their care, to Dr S. Sevitt and Dr H. Thompson for post-mortem details, to Dr P. Jacobs for radiological interpretation, to Dr J. F. Riordan for performing lung function studies in Case 2 and to Mr N. R. Gill for photographic reproduction. REFERENCES Ashbaugh D. C., Petty T. L., Bigelow D. B. et al. (1969) Continuous positive-pressure breathing (CPPB) in adult respiratory distress syndrome. J. Thorac. Cardiovasc. Sup. 57, 3 1. Benzinger T. (1950) Physiological effects of blast in air and water. In: German Aviation Medicine World War II, vol. 2. Washington, US Government Printing Office, p. 1225. de Candole C. A. (1967) Blast injury. Can. Med. Assoc. J. 96, 207.
Carlton L. M. jun., Rasmussen R. A. and Adams W. E. (1945) Blast injury of lung: possible explanation of mechanism in fatal cases--experimental study. Surgery 17, 786. Cohn J. N. (1967) Central venous pressure as a guide to volume expansion. Ann. Intern. Med. 66, 1283. Fraser R. G. and Pare J. A. P. (1970) Diagnosis of Diseases of the Chest, vol. 2. Philadelphia, Saunders, p. 1069. Gray R. C. and Coppel D. L. (1975) Surgery of violence. III. Intensive care of patients with bomb blast and gunshot injuries. Br. Med. J. 1,502. Hadfield G. and Christie R. V. (1941) A case of pulmonary concussion (‘blast’) due to high explosive Br. Med. J. 1, 77.
Hadfield G., Ross J. M., Swain R. H. A. et al. (1940) Blast from high explosive. Preliminary report on ten fatal cases. Lancet 2, 478. Hamit H. F. (1973) Primary blast injuries. Ind. Med. Surg. 42, No. 3, 14.
Hirsch M. and Bazini J. (1969) Blast injury to thechest. Clin. Radiol. 20, 362.
Hubav C. A.. Waltz R. C.. Brecher G. A. et al. (1954) Circulatory dynamics ‘of venous return during positive-negative pressure respiration. Anesthesiology 15, 445.
Huller T. and Bazini J. (1970) Blast injuries of the chest and abdomen. Arch. Surg. 100,24. Jackson D. M. (1975) The sick cell syndrome in burns. In: Vrabec R., Konickova Z. and Moserova J. (ed.) Basic Problems in Burns. Prague, Avicenum, p. 174. Kelman G. R., Nunn J. F., Prys-Roberts C. et al. (1967) The influence of cardiac output on arterial oxygenation: a theoretical study. Br. J. Anaesth. 39, 450.
Kumar A., Falke K. J., Gefhn B. et al. (1970) Continuous positive-pressure ventilation in acute respiratory failure. Effects on hemodynamics and lung function. N. Engl. J. Med. 283, 1430. McCaughey W., Coppel D. L. and Dundee J. W. (1973) Blast injuries to the lungs. A report of two cases. Anaesthesia 28, 2. McGrigor D. B. and Samuel E. (1945) The radiology of war injuries. Part Ill. War wounds of the chest. Br. J. Radiol. 18,133. Maloney J. V. jun., Affeldt J. E., Sarnoff S. J. et al. (1951) Electrophrenic respiration. IX. Comparison of effects of positive-pressure breathing and electrophrenic respiration on the circulation during haemorrhagic shock and barbiturate poisoning. Surg. Gynecol.
Obstet.
92, 672.
Obdrzalek J., Kay J. C. and Noble W. H. (1975) The effects of continuous positive-pressure ventilation on pulmonary oedema, gas exchange and lung mechanics. Can. Anaesth. Sot. J. 22, 399. O’Reilly J. N. and Gloyne S. R. (1941) Blast injury of the lungs. Lancer 2, 423. Pontoppidan H., Laver M. B. and Geffin B. (1970) Acute respiratory failure in the surgical patient. In: Welch C. (ed.) Advances in Surgery, vol. 4. Chicago, Year Book, p. 163. Powers S. R. jun., Mannal R., Neclerio M. et al. (1973) Physiologic consequences of positive endexpiratory pressure (PEEP) ventilation. Ann. Surg. 178, 265.
Qvist J., Pontoppidan H., Wilson R. S. et al. (1975) Hemodynamic responses to mechanical ventilation with PEEP: the effect of hypervolemia. Anesthesioloav 42. 45.
RiissG R.’ (1950) Pathology German
Aviation
Medicine
of blast World
effects.
War II, vol.
In: 2.
Washington, US Government Printing Office, p. 1260. Shires G. T., Carrico C. J. and Canizaro P. C. (1973) Shock : Major Problems in Clinical Surgery, vol. 13. Philadelphia, Saunders, p. 88.
12
Injury:the British Journal
Stevens E. and Templeton A. W. (1965) Traumatic non-penetrating lung contusion. Radiology 85, 247. Suter P. M., Fairley H. B. and lsenberg M. D. (1975) Optimum end-expiratory airway pressure in patients with acute pulmonary failure. N. Engl. J. Med. 292, 284. Sykes M. K. (1963) Venous pressure as a clinical indication of adequacy of transfusion. Ann. R. Co//. Surg. Engl. 33, 185. Sykes M. K., McNicol M. W. and Campbell E. J. M. (1969) Respirafory Failure. Oxford, Blackwell Scientific, p. 215. Waterworth T. A. and Carr M. J. T. (1975) An analysis of the post-mortem findings in the 21 victims of the Birmingham pub bombings. lnury 7, 89.
Requests for reprints should be addressed to: Dr N. Queen Elizabeth Hospital, Birmingham, 815 2TH.
G.
Caseby,
of Accident
Surgery
Vol. ~/NO. 1
Weiler-Ravel1 D., Adatto R. and Borman J. B. (1975) Blast injury of the chest. A review of the problem and its treatment. Isr. J. Med. Sri. 11, 268. White C. S., Bowen 1. G. and Richmond D. R. (1964) A comparative analysis of some of the immediate environmental effects at Hiroshima and Nagasaki. Health Phys. 10, 89. Williams J. R. and Stembridge V. A. (1964) Pulmonary contusion secondary to non-penetrating chest trauma. Am. J. Roentgenol. Radium Ther. Nucl. Med. 91, 284. Wilson J. V. (1943) The pathology of closed injuries of the chest. Br. Med. J. 1, 470. Wilson J. V. and Tunbridge R. E. (1943) Pathological findings in a series of blast injuries. Lancer 1, 257. Zuckerman S. (1940) Experimental study of blast injuries to the lungs. Lancet 2, 219.
FFA
RCS,
Department
of Anaesthetics,
University
of‘ Rirmingham,