Fires cause burns and these injuries are obvious but injuries to the lungs and airways from smoke inhalation are often less apparent and may not present until 24 to 36 hours after exposure.¹ Smoke injury accounts for 75% of the deaths related to fire, although there may be burns too. Another major cause of serious injury and death related to fires is people jumping from high windows rather than awaiting the arrival of the emergency services.

Burns and smoke inhalation are often inter-related in that the greater the surface area burned, the more likely there is to be smoke injury too. They both contribute to morbidity and mortality and should be assumed to coexist.²

Epidemiology^2,3

A study from North West England found that the overall admission rate to hospital for smoke inhalation or burns was 0.29/1,000 population per year. Another found that 10 to 30% of all burns admissions had smoke inhalation injury. The groups most at risk are the under-5s and the over-75s, with a male to female preponderance of about 2:1. Mortality is highest in the elderly and risk increases with social deprivation.

Risk factors¹

Risks are increased by being in a confined space, by the duration of exposure, by substances being burned that may emit various poisons and by pre-existing respiratory disease.

Mechanism of injury⁴

There are three basic ways that damage occurs. Heat causes thermal damage, gases cause asphyxiation and there may be irritation of the lungs or airways.

Generally, heat damage is limited to the oropharynx, as heat is soon dissipated, but steam and explosions may carry it rather deeper.

Asphyxiation results in tissue hypoxia. The fire may take a significant amount of oxygen from the air leaving only 10 -13% oxygen, giving a pO₂ of barely half the normal value. Even small amounts of carbon monoxide (CO) will aggravate the situation, as the gas has 200 times the affinity of oxygen for haemoglobin so that an atmosphere of 21% oxygen and 0.1% CO will cause the blood to leave the lungs 50% saturated with oxyhaemoglobin and 50% saturated with carboxyhaemoglobin.

CO also binds to myocardial myoglobin and reduces the contractility of the heart. Combustion of plastics, polyurethane, wool, silk, nylon, nitriles, rubber and paper products can all lead to cyanide production. Only about 40% of the population can smell hydrogen cyanide as an almond-like odour. It is 20 times as toxic as CO and can cause immediate respiratory arrest.⁴ If patients with smoke inhalation have cardiovascular complications then cyanide poisoning should be considered.

Irritation of the lungs and airways will incite an inflammatory response with bronchospasm and an outpouring of fluid. This seems very much like the inflammatory response in acute asthma but steroids are of no benefit.⁵ Chlorine and oxides of sulphur and nitrogen cause acids, whilst ammonia is alkaline. Highly soluble substances tend to aggravate the upper airways but less soluble poisons like phosgene and nitrogen oxides reach the lungs. Airways can be blocked by soot particles along with bronchospasm and the outpouring of inflammatory exudate. Pulmonary oedema may take a while to develop and so observation after smoke injury is important.

General assessment⁴

• Look at the patient. Is breathing normal? Is it laboured? Is there cyanosis? Does the chest wall move normally and symmetrically?
• Assess the airway but, if there is any risk of cervical spine trauma, be careful with the neck.
• Note respiratory rate. Listen to the chest.
• Note level of consciousness, pulse rate, blood pressure and peripheral circulation.
• If the patient is not fully conscious and alert the Glasgow Coma Scale should be employed.
• Note any injuries and burns, undressing the patient as required and possibly removing smoldering clothes. Check the back too.
• Respiratory assessment is required in anyone with possible smoke injury. It may form a baseline, as conditions can deteriorate after rescue.
• Check for signs of deteriorating respiratory function and treat aggressively before the situation becomes desperate.
• Hoarseness and change in the voice may herald serious problems and tachypnoea is a bad sign.
• Black sputum suggests excessive exposure to soot .
• Note rhonchi, rales and use of accessory muscles of respiration.
• Facial burns show nearness to the fire. Other burns demonstrate an inability to escape.

Investigations^1,4

• A baseline chest X-ray (CXR) may be useful for comparison if pulmonary oedema ensues.⁶ Early CXR is often normal and a normal film should not give too much reassurance. Later features can include atelectasis, pulmonary oedema and acute respiratory distress syndrome.
• Blood gases should be performed, including carboxyhaemoglobin and acid/base balance.
• A pulse oximeter may give false readings by interpreting carboxyhaemoglobin as oxyhaemoglobin. Cooximetry, a 4-wavelength technique of light refractance to accurately measure carboxyhaemoglobin and oxyhemoglobin, gives a more accurate assessment.
• U&E and creatinine as a baseline are also important if there are substantial burns or crush injuries.
• ECG may show evidence of cardiac ischaemia, especially after cyanide exposure .
• Bronchoscopy may be very useful in identifying erythema, oedema, ulceration, the presence of carbonaceous material, and atelectasis.
• Radionucleotide scintigraphy may be helpful if bronchoscopy is normal or borderline but the clinical picture suggests smoke inhalation.
• Cyanide may be emitted in household fires. An elevated plasma lactate may point to this.⁷

Management^1,4

• Immediate management at the scene involves extracting the patient as rapidly as is safe, bearing in mind other possible injuries, and getting out into the fresh air. Then (and only once clear of the fire!) oxygen should be given at high flow rate and humidified. Establish venous access, assess briefly and then transport with the minimum of delay. The most experienced people at dealing with smoke injury are in a burns unit and there may well be burns too.
• A patient who has suffered smoke inhalation should be assumed to have CO poisoning and be treated accordingly. High flow 100% O₂ reduces the half life of CO in the blood from over five hours in normal air to about 75 minutes.⁸ CO poisoning is responsible for most of the deaths which occur before reaching hospital. If it causes cardiac arrest, the chance of resuscitation is extremely poor. Hyperbaric oxygen at 3 bars reduces the half life of CO down to around 23 minutes. This may be very valuable in preventing neurological damage that tends to be a reperfusion injury.
• About 50% of patients with smoke injury will need intubation. This is more likely if there are also burns. If there is a significant risk of intubation being required it should be performed early or else oedema may make it technically more difficult or impossible. Damage to the mucosa of the trachea makes it more vulnerable and so the endotracheal tube cuff should not be overinflated. Even allow a little leakage.
• In hospital, intubation and ventilation may be required. Positive end expiratory pressure (PEEP) gives the best results.⁹ High-frequency ventilation has been found to be helpful.¹⁰ One study found that there was a strong association between laryngeal oedema (and hence the need for intubation) and facial burns, body burns and/or soot in the oral cavity.¹¹
• Intravenous fluids are required to prevent dehydration and to assure adequate perfusion. If there are burns too the fluid requirements should be related to this.
• The treatment of cyanide poisoning is with amyl and sodium nitrite that create methaemoglobinaemia. This can be dangerous when there is carbon monoxide poisoning too and it should be used with caution.
• Patients with acute asthma or chronic obstructive pulmonary disease may develop bronchospasm that needs treating in the usual way.

Admission policy⁴

Patients who have suffered smoke inhalation but are not definite candidates for admission should be monitored in A&E for 4 to 6 hours before discharge. The following point to the need for admission:

• Exposure in a closed space for more than 10 minutes
• Thick, black sputum
• PaO₂ below 8 kPa (60 mmHg) or metabolic acidosis
• Carboxyhaemoglobin above 15%
• Arteriovenous oxygen difference (on 100% oxygen) greater than 13.33 kPa (100 mmHg)
• Bronchospasm
• Burns to the face

Complications⁴

• One American study looking at children aged 10-18 years, admitted with inhalation injury over a 10-year period (a sample of 850), reported a mortality rate of 16.4%.¹² Patients with burns as well as smoke injury have up to four times the mortality.⁹ This may be in part because of more severe exposure but, the presence of burns with smoke injury, means that the patient has to be treated more seriously.
• Subglottic stenosis, bronchiectasis, pulmonary oedema and atelectasis can occur.
• It seems quite possible that smoke injury may trigger susceptibility to asthma.¹³ Polyvinyl chloride (PVC) is a particular risk and post-exposure treatment with non-steroidal anti-inflammatory drugs may help.¹⁴
• Do not rely on CO levels in A&E to assess severity of exposure. The level at the scene may have been more relevant but beware of complications taking many hours to arrive.
• If a patient is discharged they must be given written instructions on the warning signs which would indicate the need to return immediately. Those with normal vital signs and examination, and with short exposure may safely be discharged.¹⁵ The experience from disaster triage involving smoke inhalation is that patients without dermal burns and with normal bronchoscopy, chest radiography and normal blood gases (including carboxyhaemoglobin levels) have a low risk of complications.¹⁶

Prognosis

Due to its relative rarity, information on prognosis is not abundant. However, in one series of 96 patients, 13 patients developed immediate respiratory failure resulting from ventilatory insufficiency, 4 patients died, and vocal cord and tracheal stenosis were noted in 5 patients and 1 patient respectively. In patients with reduction in pulmonary function, improvement was seen after three months, with no further changes being observed within the subsequent three months.¹⁷ An American review quoted an increased mortality by a maximum of 20% over that predicted by age and extent of cutaneous burn alone. Pneumonia in these patients further increased mortality by a maximum of 40%. One estimate suggested that 75% of deaths following burn injury could be accounted for by inhalation injury but more recent cohort studies have suggested a decreasing mortality attributable to this cause.²

Prevention

The prevention of smoke injury is largely the prevention of fire but, if it does occur, then early warning is necessary. Smoke detectors save lives, perhaps reducing fatalities by up to 60% but they only work if an effective battery is in situ and many people are lax about checking this. Even those less likely to respond so swiftly to an alarm, like the very young, the elderly, the infirm and those intoxicated by drugs or alcohol, may benefit.¹⁸ Programmes to give away smoke alarms have not been randomised and American experience suggests that the batteries are not kept in order.¹⁹ One study suggests that the use of an alarm which plays a recording of the parent's voice is more likely to wake a child than a conventional residential tone smoke alarm.²⁰ Initiatives to increase the uptake of alarms, such as incorporating them into child surveillance programmes, require further evaluation. Alarms which have a low 'nuisance' level (e.g. which do not sound unnecessarily) seem to provide the most effective prevention.²¹

Deaths in children under five are sometimes associated with fireplay and these are not usually prevented by smoke alarms, due to the behaviour of the children. Interventions to prevent fireplay in this age group may be more successful.²²

The choice of household furnishings is important in terms of risk of emission of toxic gases on burning as well as combustibility. There are relevant laws about materials that may be used in the manufacture of furniture.

Document references

1. Serebrisky D, Nazarian E; Inhalation Injury eMedicine.com 2008.
2. Sterner JB, Zanders TB, Morris MJ, et al; Inflammatory mediators in smoke inhalation injury. Inflamm Allergy Drug Targets. 2009 Mar;8(1):63-9. [abstract]
3. Rajpura A; The epidemiology of burns and smoke inhalation in secondary care: a population-based study covering Lancashire and South Cumbria. Burns. 2002 Mar;28(2):121-30. [abstract]
4. Lafferty K; Smoke Inhalation eMedicine.com 2008.
5. Nieman GF, Clark WR, Hakim T; Methylprednisolone does not protect the lung from inhalation injury. Burns. 1991 Oct;17(5):384-90. [abstract]
6. Saab M, Majid I; Acute pulmonary oedema following smoke inhalation. Int J Clin Pract. 2000 Mar;54(2):115-6. [abstract]
7. Baud FJ, Barriot P, Toffis V, et al; Elevated blood cyanide concentrations in victims of smoke inhalation. N Engl J Med. 1991 Dec 19;325(25):1761-6. [abstract]
8. Weaver LK, Howe S, Hopkins R, et al; Carboxyhemoglobin half-life in carbon monoxide-poisoned patients treated with 100% oxygen at atmospheric pressure. Chest. 2000 Mar;117(3):801-8. [abstract]
9. Clark WR Jr; Smoke inhalation: diagnosis and treatment. World J Surg. 1992 Jan-Feb;16(1):24-9. [abstract]
10. Cancio LC; Airway management and smoke inhalation injury in the burn patient. Clin Plast Surg. 2009 Oct;36(4):555-67. [abstract]
11. Madnani DD, Steele NP, de Vries E; Factors that predict the need for intubation in patients with smoke inhalation injury. Ear Nose Throat J. 2006 Apr;85(4):278-80. [abstract]
12. Palmieri TL, Warner P, Mlcak RP, et al; Inhalation injury in children: a 10 year experience at Shriners Hospitals for J Burn Care Res. 2009 Jan-Feb;30(1):206-8. [abstract]
13. Park GY, Park JW, Jeong DH, et al; Prolonged airway and systemic inflammatory reactions after smoke inhalation. Chest. 2003 Feb;123(2):475-80. [abstract]
14. Moisan TC; Prolonged asthma after smoke inhalation: a report of three cases and a review of previous reports. J Occup Med. 1991 Apr;33(4):458-61. [abstract]
15. Mushtaq F, Graham CA; Discharge from the accident and emergency department after smoke inhalation: influence of clinical factors and emergency investigations. Eur J Emerg Med. 2004 Jun;11(3):141-4. [abstract]
16. Goh SH, Tiah L, Lim HC, et al; Disaster preparedness: Experience from a smoke inhalation mass casualty incident. Eur J Emerg Med. 2006 Dec;13(6):330-4. [abstract]
17. Cha SI, Kim CH, Lee JH, et al; Isolated smoke inhalation injuries: acute respiratory dysfunction, clinical Burns. 2007 Mar;33(2):200-8. Epub 2006 Dec 13. [abstract]
18. DiGuiseppi C, Higgins JP; Interventions for promoting smoke alarm ownership and function. Cochrane Database Syst Rev. 2001;(2):CD002246. [abstract]
19. Marshall SW, Runyan CW, Bangdiwala SI, et al; Fatal residential fires: who dies and who survives? JAMA. 1998 May 27;279(20):1633-7. [abstract]
20. Smith GA, Splaingard M, Hayes JR, et al; Comparison of a personalized parent voice smoke alarm with a conventional residential tone smoke alarm for awakening children. Pediatrics. 2006 Oct;118(4):1623-32. [abstract]
21. Mueller BA, Sidman EA, Alter H, et al; Randomized controlled trial of ionization and photoelectric smoke alarm Inj Prev. 2008 Apr;14(2):80-6. [abstract]
22. Istre GR, McCoy M, Carlin DK, et al; Residential fire related deaths and injuries among children: fireplay, smoke alarms, and prevention. Inj Prev. 2002 Jun;8(2):128-32. [abstract]

Internet and further reading

• A Guide to the Furniture and Furnishings (Fire) (Safety) Regulations; Department of Trade and Industry 2007
• Smoke Alarms; Merseyside Fire & Rescue Service 2007
• Smoke inhalation, Health Protection Agency

Acknowledgements