Traditionally drowning has been defined as death due to suffocation within 24 hours of submersion in a liquid medium and near-drowning as survival for 24 hours or more following such an incident.¹ Considerable confusion has surrounded the use of these terms. In part this is because the distinction between drowning and near-drowning often cannot be made before 24 hours, making the terms clinically irrelevant. In addition it has been suggested that the use of a time limit for survival is not a scientific concept and is not in accordance with outcome parameters as used in the internationally accepted Utstein style.¹

To address this issue the Utstein Taskforce on Drowning was convened in Amsterdam as part of the 2002 World Congress on Drowning. The International Liaison Committee on Resuscitation (ILCOR) has since endorsed a review of the terminology and defines drowning in the following way:

Drowning. Drowning is a process resulting in primary respiratory impairment from submersion/immersion in a liquid medium. Implicit in this definition is that a liquid/air interface is present at the entrance of the victim’s airway, preventing the victim from breathing air. The victim may live or die after this process, but whatever the outcome, he or she has been involved in a drowning incident. ²

The term submersion is generally accepted to indicate an incident in which the victim’s body is totally covered by water, while the term immersion refers to an incident in which the victim is only partially covered by water, although for drowning to occur the face and airway must at least be covered.²

ILCOR recommends that other terms such as dry drowning versus wet drowning, active versus passive versus silent drowning, secondary drowning and near drowning be abandoned.²

Epidemiology

In Australia drowning is a leading cause of accidental death in children. Its incidence peaks in early childhood and again in adolescence. Males outnumber females in both groups. Children under five years of age are the most vulnerable to drowning in Australia.³ In the period between 1 July 2008 and 30 June 2009 there were 302 drowning deaths in Australian waterways. Children aged 0-4 years accounted for 11% of deaths overall and 74% of deaths under the age of 14 years. 59% of drowning deaths in the 0–4 years age group occurred in swimming pools with around 84% of cases occurring from wandering or falling in.⁴

Risk groups for childhood drowning are children aged 0–4 years, children living in cities with high swimming pool to population ratios, children living in hot climates, children living in areas with lack of isolation pool fencing, and Indigenous children.⁵ More toddlers drown in swimming pools than from any other cause.⁶ Most children who drown in pools are out of sight for less than 5 minutes and are in the care of one or both parents.⁷ Around the home small children can also drown in baths, buckets, and garden ponds. Up to 8% of cases of drowning in small children in the domestic setting may be secondary to non-accidental injury.⁸

While pool-fencing legislation has proven to be effective in reducing the incidence of drowning in small children it has had little impact on rates of drowning in older children and adolescents. In this group alcohol, suicide, and risk-taking behaviours are important factors that lead to increased risk of drowning.⁸

Aetiology

Drowning is most commonly a primary event. In children it most often occurs when the victim is unable to rescue him or herself after entering the water, as in the case of a toddler falling into a swimming pool, or an infant drowning in a bath whilst unattended. In older children and adolescents fatigue while swimming may play a role, but drowning in these age groups is more likely to be secondary to other causes.

Drowning can occur secondarily to a number of underlying causes. These should be considered during assessment of the submersion victim. Individuals with seizure disorders have up to 19-times higher risk for drowning accidents, regardless of age.^3,^8,⁹ Prolonged QT-syndrome leading to dysrhythmia has been implicated as a significant cause of drowning, although the true incidence of this condition in drowned children is unknown. Ethanol is an important risk factor for drowning injury, particularly in adolescents. Elevated serum ethanol levels are documented in 10–50% of adolescent drownings.⁸ Head and cervical trauma from diving and boating-related accidents may also lead to drowning as a secondary event. Non-accidental injury is an important cause of drowning in infants and smaller children, particularly in events that occur in the home, such as in baths and buckets. Up to 8% of drownings presenting to tertiary paediatric centres may be attributed to child abuse.⁸

The immersion syndrome is sudden loss of consciousness secondary to a bradycardia, or tachyarrhythmia induced by contact with water at a temperature of at least 5°C below body temperature. This can lead secondarily to drowning. The immersion syndrome can occur in water with temperatures as warm as 31°C, although it is more likely to occur in much colder water. Wetting the face before entering the water may reduce its incidence.¹

Pathophysiology

The two most significant pathophysiological consequences of submersion are hypoxia from asphyxiation during the submersion itself, and aspiration of water into the lungs. It is the severity of the initial hypoxic insult that is the major determinant of outcome. If the initial hypoxic event is survived, the degree of hypoxic organ injury and pulmonary injury secondary to aspiration become the clinically important factors.

Much that is known about the sequence of events following submersion has come from animal models. Aspiration of water initially causes breath-holding or laryngospasm and the resultant asphyxiation leads to progressive hypoxia. Active and passive swallowing of water follows and, as hypoxia worsens, breath-holding and laryngospasm are terminated, resulting in aspiration of water into the lungs.¹

Anoxia lasting 1–3 minutes can shut down both the brain and the heart, causing loss of consciousness and hypoxic cardiac arrest. Rescue and early institution of cardiopulmonary resuscitation can salvage myocardial function, but the brain is more sensitive to hypoxic injury and it is the severity of this injury that determines outcome. Effects of hypoxia on other organ systems are delayed. Profound hypoxia can cause an acute respiratory distress syndrome, which develops within hours and further worsens hypoxic injury. Posthypoxic cerebral oedema is a major complication and can develop 6–12 hours following successful initial resuscitation from a serious submersion event. Most paediatric drowning deaths in hospital are due to hypoxic cerebral injury rather than pulmonary complications.⁸

The average volume of water aspirated in human drownings is 10–15 mL kg^–1. Aspiration of volumes as little as 1–3 mL kg^–1 of water can cause profound alterations in gas exchange and subsequent ventilatory abnormalities.¹⁰ Laryngospasm is thought to occur in 10–15% of drowning victims and a subset of patients who drown without evidence of significant aspiration of water at post-mortem, so-called dry-drowning, has been described. This concept has recently been questioned and it has been suggested that in these cases death has occurred prior to submersion.¹ Regardless of whether dry-drowning is a true clinical entity, or whether laryngospasm has occurred at the time of submersion, aspiration of water into the lungs remains a clinically important consideration in the management of all drowning victims.

Despite the large literature dedicated to the subject, there are no clinically or therapeutically important differences between drowning in fresh or salt water.^1,⁸ Pulmonary injury is related more to the amount of water aspirated than to the composition of the water itself. Both fresh and salt water cause loss of pulmonary surfactant, non-cardiogenic pulmonary oedema, impaired alveolar-capillary gas exchange, and increased intrapulmonary shunting with the potential for profound hypoxia.¹ Aspiration of water that is contaminated with particulate matter or bacteria can lead to complications from obstruction of small airways or increased risk of pulmonary infection, although neither is seen in the majority of patients.¹¹ If present, evidence of significant pulmonary injury due to aspiration will usually manifest or progress within hours of rescue. Delayed onset of respiratory distress and hypoxia, the so-called ‘delayed immersion syndrome’ or ‘secondary drowning’ has been refuted by recent evidence.⁸

Clinically significant electrolyte and fluid volume abnormalities are rarely seen in cases of drowning in humans despite being demonstrated in animal models.^1,^3,^8,¹¹ Occasionally a mild hyponatraemia, which self-corrects without specific therapy, is observed.⁸ Theoretical exceptions are drownings occurring in hypertonic solutions, such as the Dead Sea, or water contaminated with industrial waste.

Hypothermia is an important issue following drowning, particularly in small children who have a large body surface area to weight ratio. Cooling can occur at the time of submersion, but can also continue following rescue and during attempted resuscitation due to heat loss through evaporation. Hypothermia can confer some degree of protection from cerebral hypoxia, particularly in small children. Multiple case reports in the literature attest to intact survival of both children and adults following prolonged (>15 minutes) drownings in icy water (water temperature <10°C).¹² Profound hypothermia and subsequent intact survival has also been documented in children suffering drowning in non-icy water and in temperate climates.

The mechanisms of temperature drop and cerebral protection remain unclear. Surface cooling at the time of submersion is thought to be insufficient on its own to provide central cooling of a degree that confers cerebral protection. Other mechanisms of heat loss, such as via ingestion and/or aspiration of cold water, are not supported by quantitative evidence. Some authors suggest that core temperature drop is insufficient on its own to explain the cerebral protection afforded by hypothermia.¹³ The diving reflex, in which blood is shunted from the limbs and splanchnic circulation to the brain and heart alongside slowing of the heart rate and reduction of the basal metabolic rate, has been suggested as being an important mechanism for cerebral protection in children.¹³ There is little clinical evidence to indicate that the diving reflex is sufficiently active in humans, even small children, to confer any benefit on its own.^1,⁸ It is most likely that a combination of the effects of the diving reflex initially, followed by rapid and continued cooling, is what underlies the cerebral preservation that is sometimes seen in small children who suffer submersion and who are profoundly hypothermic.

By whatever mechanism cooling occurs, and whether the diving reflex plays a significant role in cerebral protection or not, hypothermia in the drowning victim, particularly if the victim is an infant or young child, should be considered to be an indication for aggressive and prolonged resuscitation efforts. This issue is discussed further in Chapter 22.4 on cold injuries.

History

Key points in history are summarised in Table 22.2.1. Broad areas in history include details of the drowning event itself, details of rescue and resuscitation, response to resuscitative efforts, possible underlying causal factors, and medical conditions that may influence recovery. Features of history that alert to the possibility of non-accidental injury should be recognised. These include a history that is inconsistent or a history that is incompatible with the victim’s developmental level.

The circumstances leading to the submersion The location of the submersion Water and environmental temperatures Duration of submersion Time from when the victim was last seen to the time when found (if the submersion was unwitnessed) Time from rescue to effective CPR Time to first gasp or return of spontaneous circulation Elements of basic and advanced life support employed Symptomatology following successful resuscitation Medical history particularly with regard to possible predisposing factors (i.e. neurological disorders, such as epilepsy) Factors that may complicate recovery (i.e. respiratory illness, such as asthma)

Examination

Examination is dictated by the clinical condition of the victim on arrival in the emergency department (ED) and is mainly directed at assessing the degree of neurological impairment due to hypoxic cerebral injury, the severity of respiratory embarrassment due to aspiration, and cardiovascular instability due to a combination of the initial insult and/or ongoing hypoxia. In broad terms drowning victims arriving in the ED will fall into two categories: (1) those who respond to minimal resuscitation and who will generally do well with minimal complications; and (2) those ‘high risk’ patients who fail to respond to resuscitation and who will require ongoing resuscitation and/or monitoring.⁸

Vital signs, including oxygen saturation, a bedside blood glucose estimate, and a temperature should be recorded on all patients. It is important to detect hypothermia, particularly in patients who have failed to respond to resuscitation efforts. Small children may rapidly become hypothermic due to evaporative heat loss during resuscitation efforts and transfer to the hospital. Apparent lifelessness and severe bradycardia due to profound hypothermia need to be distinguished from asystole and brain death. Assessment of cardiac rhythm requires observation of the continuous electrocardiograph (ECG) monitor for up to a minute to detect very slow heart rates that can occur in profound hypothermia. Similarly, lack of response to painful stimulus, along with fixed and dilated pupils should not be interpreted as brain death in the presence of profound hypothermia.

Depending on the clinical condition of the victim at arrival, neurological evaluation may range from an assessment of level of responsiveness as graded by the AVPU or Glasgow coma score (GCS) and pupillary reaction, to a focused neurological examination looking for focal deficit or spinal-cord injury. An assessment should always be made for the possibility of cranial or cervical spine trauma, particularly in diving-related accidents. Trauma to other areas of the body should also be sought.

In the awake child, examination of the respiratory system can establish a clinical baseline against which subsequent deterioration can be measured. Work of breathing should be assessed, and abnormalities found on auscultation, such as crackles and wheezes, should be noted. The finding of signs on an initial examination indicates the possibility of significant aspiration and dictates close observation to identify deterioration. A clear chest initially does not exclude aspiration and a period of observation and re-examination is necessary in all children who have been symptomatic following drowning.

Investigations

Investigations will also be dictated by the clinical condition of the patient on arrival in the ED. A child who has suffered a drowning injury and who is alert and asymptomatic on arrival requires little in the way of laboratory or radiological evaluation. The more symptomatic or seriously unwell child may benefit from further evaluation with laboratory and radiological investigations.

Useful investigations include arterial blood gas (ABG) analysis, serum glucose, and chest radiography. ABG analysis may demonstrate a metabolic acidosis, which confirms a significant drowning injury.⁸ The severity of the acidosis reflects the severity of the hypoxic insult as well as ongoing hypoxic injury. Profound acidosis (pH <7.10) implies a poorer prognosis but needs to be interpreted in the clinical context.¹⁴ ABG analysis can be used to guide decisions regarding oxygenation and ventilation, and serial determinations may be useful in monitoring and quantifying deterioration of pulmonary function due to non-cardiogenic pulmonary oedema secondary to hypoxia or aspiration. Determination of the blood glucose level is important in any critically ill child as hypoglycaemia can complicate physiological stress and should be actively treated. Hyperglycaemia in a comatose child, although not requiring treatment, implies a poor prognosis.⁸ Initially the chest X-ray (CXR) may be normal, may demonstrate pulmonary infiltrates, or may display frank pulmonary oedema. Abnormalities detected on an early CXR mandate close observation and the patient should be monitored for clinical deterioration. Repeated CXR may be required but should be dictated by the patient’s clinical condition.

Other investigations that may be helpful include baseline electrolytes and full blood count, although rarely are any clinically or therapeutically significant abnormalities found on initial determinations. Blood ethanol levels may be relevant, depending on the age of the patient. A 12-lead ECG may be helpful in excluding prolonged QT syndrome as a cause for the drowning event.¹⁰ Cervical spine films should be considered if cervical injury is suspected or the drowning is secondary to a diving accident.

Differential diagnosis

The major issues in differential diagnosis relate to the cause of the drowning event. Underlying medical conditions, such as epilepsy, should be considered. Trauma, either leading to or as a consequence of the drowning, should be recognised. Non-accidental injury should be suspected when there are inconsistencies in the history.

Severe hypothermia (core temperature less than 29°C) can mimic irretrievable cardiorespiratory arrest and brain death. The profound bradycardia associated with very low core body temperature can easily be confused with brady-asystole secondary to hypoxic injury and will not respond to usual resuscitation measures until hypothermia is corrected. Similarly, severe hypothermia can cause depression of cerebral function leading to unresponsiveness and fixed, dilated pupils, indistinguishable from irreversible hypoxic cerebral injury.¹⁵ Failure of the core temperature to rise despite aggressive, active rewarming may be the only indication that death has occurred.

Treatment

Treatment of the drowning victim occurs in three major phases: (1) rescue and resuscitation at the scene; (2) initial assessment and stabilisation in the emergency department; and (3) subsequent observation and supportive care in the hospital ward, intensive-care unit (ICU), or after discharge.

Early institution of effective cardiopulmonary resuscitation with an emphasis on providing adequate ventilation is the key task in pre-hospital management. Manoeuvres to drain the lungs of water have not been shown to be clinically effective and may increase the risk of aspiration of gastric contents. The Heimlich manoeuvre should be reserved for cases in which repeated attempts to position the airway and provide ventilation have been unsuccessful or when foreign body aspiration and obstruction is suspected.¹ Emesis is common in drowning victims, both spontaneously and as a complication of resuscitation, and aspiration of gastric contents is a major potential complication following rescue. Spontaneously breathing patients should be managed and transported in the right lateral decubitus position. Cricoid pressure may reduce the risk of gastric distension and aspiration during cardiopulmonary resuscitation but requires an additional rescuer.¹

Hypothermia can be exacerbated by ongoing evaporative heat loss during resuscitation efforts following rescue. Wet clothing should be removed and the patient should be dried if possible. Exposure during cardiopulmonary resuscitation (CPR) should be minimised as much as is practicable. If hypothermia is severe at the scene (<30°C), rewarming should probably be delayed until adequate ventilation and oxygenation has been instituted.¹⁵ Invasive rewarming techniques should be reserved for the hospital phase of management.

Treatment in the ED involves provision of adequate oxygenation, stabilisation of the body temperature, prevention of complications such as aspiration, and assessment of the patient’s clinical status in order to make decisions regarding ongoing management and disposition.

Supplemental oxygen should be administered at concentrations and via delivery systems appropriate to the patient’s oxygen requirement. Continuous pulse oximetry and serial estimations of respiratory rate and work of breathing should be undertaken during observation in the ED to determine worsening or improvement of the patient’s respiratory status. If available, non-invasive ventilation techniques, although not extensively evaluated in paediatric drowning victims, may be considered if the oxygen requirement outstrips conventional mechanisms of oxygen delivery. Intubation to isolate the airway and provide ventilation with positive end-expiratory pressure (PEEP) may be required in patients with depressed neurological status or if respiratory compromise progresses despite supplemental oxygen therapy. Failure to maintain an SaO₂ of greater than 90% with an FiO₂ of 0.50 or higher, a PaCO₂ of more than 35 mmHg, an abnormally high respiratory rate (>50 bpm) or inadequate spontaneous ventilation, are all indications that mechanical ventilation is likely to be required.^1,¹⁰ Passage of a nasogastric or orogastric tube, if cranial trauma is suspected, should be performed following intubation to decompress the stomach and facilitate mechanical ventilation.

Diuretics are not recommended in the management of non-cardiogenic pulmonary oedema.^1,^8,¹⁰ Steroids have not been shown to be useful in the management of aspiration pneumonitis and the role of antibiotics in this setting remains controversial.^1,^8,¹⁰ Prophylactic antibiotics are generally not recommended, although they are sometimes used when there is a history of drowning in heavily contaminated or polluted water. In general, antibiotics should be reserved for patients who develop signs of pulmonary infection, such as fever or leukocytosis.¹ Fluid therapy should be judicious and should be aimed at maintaining adequate circulatory status and euglycaemia without overloading the patient.

Hypothermic patients should be warmed once adequate ventilation and oxygenation has been assured. Awake and spontaneously breathing patients with mild to moderate hypothermia (core temperature >32°C) require passive or external active rewarming techniques only. Patients with severe hypothermia (core temperature <32°C) who are obtunded or in cardiorespiratory arrest will require invasive active rewarming techniques. These are discussed fully elsewhere in this text (see Chapter 22.4 on cold injuries). Whilst there is increasing support for the use of therapeutic hypothermia (cooling to a core temperature of 33°C) following cardiac arrest in adults, and the technique has been used in adults following drowning, its role in the management of children remains to be established and may, in fact, be associated with harm.¹⁶

In the setting of cardiorespiratory arrest and severe hypothermia, resuscitation efforts should be continued until the core temperature has risen to around 32°C before a decision to cease resuscitation is made. Although there are no clear guidelines to determine how long efforts at rewarming should be continued, failure to effect rewarming despite maximal invasive efforts may be an indication that continuing resuscitation is futile. The decision to cease resuscitative efforts in the presence of persistent severe hypothermia should follow a multidisciplinary approach involving the emergency physician, paediatrician and/or paediatric intensivist and other members of the resuscitation team, together with the parents and family. Cardiorespiratory arrest on arrival in the ED in a patient with a core temperature above 32°C carries a uniformly poor prognosis and prolonged efforts at resuscitation are generally not indicated.

Treatment in the ICU involves provision of ventilatory support to ensure adequate oxygenation, maintenance of cardiovascular stability, minimisation of secondary brain injury due to cerebral oedema, management of hypoxic organ injury, and management of the complications of pulmonary aspiration. The use of PEEP in drowning victims ventilated in the ICU is controversial. Although PEEP may be necessary to maintain adequate oxygenation, some authorities express concern regarding the potential reduction in cerebral venous outflow and subsequent increase in intracerebral pressure associated with its use. PEEP may have unfavourable effects on cardiovascular status and may increase the risk of barotrauma in already insulted lungs.⁸ Other ventilation strategies to minimise non-cardiogenic pulmonary oedema and barotrauma, such as pressure-control ventilation with low-peak airway pressure and prolonged expiratory time, and permissive hypercapnia have been suggested, although again the trade-off is the risk in adversely influencing intracerebral pressure.⁸ Various measures to provide cerebral resuscitation and control intracranial pressure have been evaluated in the literature and have generally been found to be not helpful in influencing outcome.¹ In general, the aim should be to maintain adequate cerebral oxygenation and to minimise causes of secondary cerebral injury such as hypotension, hypercapnia, and hypo/hyperglycaemia.

Disposition

Disposition is dictated by the clinical condition of the patient in the ED, the nature of the drowning event, the need for resuscitation, and the presence or absence of other factors, such as trauma, suspicion of non-accidental injury, and underlying or complicating medical conditions. In general, all victims of drowning will require some period of observation.

The alert, otherwise healthy, patient who is asymptomatic or who has suffered only mild, transient symptoms following a brief drowning can be safely discharged if they remain well after 6–8 hours of observation.⁸ Patients with a history of drowning for longer than 1 minute, a period of cyanosis or apnoea, or who required pulmonary resuscitation, should be admitted for observation for 24 hours, or at least overnight, even if they are well in the ED. Although recent evidence discredits the idea of the ‘delayed immersion syndrome’, cases of fulminant pulmonary oedema up to 12 hours after drowning have been reported as occurring in patients who appear well and display normal chest radiography in the ED.¹ Any child discharged from the ED following a drowning event should be in the care of a reliable and responsible adult. Instructions should be given to re-present for further medical assessment in the event of any change in the child’s respiratory status.

Patients who have suffered trauma may require admission for the management of their injuries. Children with underlying cardiac or respiratory disorders, those who suffer drowning secondary to a pre-existing illness, and victims of suspected non-accidental injury also warrant inpatient evaluation.

Asymptomatic children and children with mild symptoms who are admitted for observation can be managed in a general ward environment, provided that there is the facility to increase the level of monitoring and care should it be required. Children who have required CPR, who have abnormal chest radiography or ABGs on arrival in the ED, or who have required ventilatory support, should be admitted to a high-dependency unit or ICU for observation and management.¹ Transfer to a facility that provides paediatric ICU should be considered for all such patients prior to clinical deterioration.

Prognosis

Most children who suffer drowning injury will either survive intact or die. Death occurs in 30–50% of drowning victims, most of these not surviving to treatment in the ED. A small proportion of victims will be left with severe neurological deficit, either persistent vegetative state or spastic quadriplegia. Mild learning deficits may occur in apparently intact survivors, although the extent of these and the impact on subsequent function have not been clearly quantified.⁸ In general, however, the prognosis of children who suffer drowning events and survive to hospital admission is excellent.¹⁷

Despite a large body of literature dedicated to the subject, the factors that predict prognosis in the ED following childhood drowning injury remain poorly defined. There are no prospectively validated scoring systems and most of the literature is based on retrospective data. Individual case reports of children surviving prolonged resuscitative efforts with good outcome continue to arise in the popular press and medical literature.

Factors elicited in history that influence prognosis are duration of submersion, time to institution of effective CPR, and time to first spontaneous gasp, all three reflecting the duration of the hypoxic cerebral insult. Duration of submersion can often be difficult to determine but submersion for longer than 5 minutes is associated with a poorer prognosis.¹⁴ Although time to institution of effective CPR can be similarly difficult to ascertain accurately, delays of longer than 10–20 minutes are also associated with poorer outcomes, while early effective CPR has been shown to be an important factor in improving survival after rescue.¹⁴

As already discussed, hypothermia may be protective in small children who suffer drowning, particularly, but not exclusively, if the drowning has occurred in cold or icy water. However, neither hypothermia nor water temperature can be reliably used to predict outcome.^18,¹⁹

Response to resuscitation following rescue and prior to arrival in the ED is the single most important indicator of outcome in children who have suffered drowning. Children who arrive conscious in the ED after successful resuscitation have almost universally excellent outcomes. Similarly, lack of response to early resuscitation efforts and coma on arrival in the ED are associated with a poor outcome.¹⁰ In a classification system based on the neurological status of the patient on arrival in the ED, it was discovered that less than 15% of patients who were unresponsive and flaccid to painful stimulation had intact survival. This is supported by the findings of other investigators that a GCS of 5 or less on arrival in the ED is associated with a dismal outcome, either death or severe neurological disability.^1,²⁰ Lack of pupillary response in the ED has similarly been identified as a predictor of dismal outcome.²¹ The caveat concerning the use of level of responsiveness and pupillary reaction to predict poor prognosis and withdraw treatment applies to the child who presents with severe hypothermia, which may mimic brain death. In these patients lack of response to aggressive rewarming and resuscitation becomes a surrogate indicator of poor outcome. In non-hypothermic patients who are in cardiorespiratory arrest, a lack of response to 25 minutes of effective advanced life support measures is almost universally associated with poor outcome.⁸ It should be noted, however, that studies of outcome following out-of-hospital cardiac arrest in children indicate that children who suffer cardiac arrest secondary to drowning have significantly better outcomes than those who suffer cardiac arrest from other causes.