a b s t r a c t

Introduction: The outcome of submersion victims depends on submersion duration and the availability of timely and effective resuscitation. The prognostic implication of vomiting during resuscitation of submersion victims is unclear. The study sought to determine whether vomiting during resuscitation in children treated for uninten- tional Submersion injuries adversely impacts outcome.

Methods: This was a retrospective study of unintentional submersion victims under age 18 treated at an urban tertiary-care children’s hospital from 2003-2009. Submersion and victim details were obtained from hospital, EMS, and fatality records. Outcomes studied were survival at 24 hours and condition (Favorable: good/mild im- pairment or Poor: death/severe disability) at hospital discharge. Descriptive comparisons between emesis groups (yes/no) and categorical covariates were analyzed.

Results: There were 281 victims. The median age was 3 years; 66% were males. Most incidents occurred at swim- ming pools (77%) and bathtubs (16%). Most were hospitalized (83%). The presence or absence of emesis was doc- umented in 246 (88%). Victims with emesis were significantly less likely to have apnea or be intubated in the ED, have a low ED GCS or die. No patient who had emesis died at 24 hours or had a poor outcome at hospital dis- charge. Victims who had emesis post-resuscitation were significantly more likely to have received CPR or chest compressions than rescue breaths.

Conclusions: Emesis in pediatric submersion victims is inversely associated with death at 24 hours or poor out- come at hospital discharge. The relationship between emesis and the adequacy of resuscitation of pediatric sub- mersion victims needs to be further studied.

(C) 2015

Introduction

Drowning is the second leading cause of unintentional injury death in children less than 14 years of age [1-3]. In 2012, the Centers for Dis- ease Control and Prevention (CDC) reported 708 deaths from accidental drowning in this age group in the United States; 458 of these deaths oc- curred in children under the age of five [4]. Known risk factors of drowning are male gender, young age, alcohol use, risky behavior, water exposure, low income status, lack of supervision, prior diagnosis of epilepsy and minority race/ethnicity [2,4-10]. Key factors that deter- mine patient outcome are submersion duration, bystander resuscitation and duration of resuscitation [11-13]. However, few studies have

? Source of support: None.

?? The abstract of the article was presented at the Pediatric Academic Societies meeting in Vancouver, BC in May, 2014.

* Corresponding author at: Section of Emergency Medicine, Baylor College of Medicine, 6621 Fannin, Suite A 2210, Houston, Texas. Tel.: +1 832 824 5422.

E-mail addresses: [email protected] (K.M. Farr), [email protected] (E.A. Camp), [email protected] (S. Yusuf), [email protected], [email protected]

(R.P. Shenoi).

explored the role of emesis during rescue of submersion victims; thus, its effect on prognosis is unclear.

Vomiting has been described as a negative prognostic indicator that interferes with pulmonary resuscitation efforts and promotes respirato- ry complications associated with aspiration of gastric contents [14-17]. Vomiting occurs both with effective and ineffective resuscitation efforts since chest compressions, abdominal thrusts, and rescue breaths can produce increased gastric pressures and subsequent regurgitation [17]. Rescuers may also attempt to force vomiting of swallowed water by holding the victim upside-down, though it is not recommended [9]. In a study of lifesaving resuscitations performed on victims of ocean submersions in Australia from 1973 to 1992, vomiting occurred in two-thirds of victims and resuscitation was more successful in victims who did not vomit [18]. Some experts recommend the use of nasogas- tric tubes to relieve gastric distension and reduce the risk of vomiting for improved Respiratory protection [19]. However, we hypothesize that pediatric submersion victims who vomit are less likely to have poor outcomes compared to those who do not vomit. We postulate that this could be due to an intact gag reflex if the submersion is brief, or it could be marker of successful resuscitation. The study sought to de- termine whether vomiting in children treated for unintentional sub- mersion injuries adversely impacts outcome.

http://dx.doi.org/10.1016/j.ajem.2015.01.055

0735-6757/(C) 2015

Outcome at hospital discharge not known (N = 13)

Fig. 1. Flowchart of submersion victim outcomes at 24 hours and hospital discharge.

No documentation of presence or absence of emesis (N = 35)

Submersion victim status at 24 hours (N = 246) Alive: 240; Dead: 6

Submersion victims (2003-2009)

N = 281

Dead

N = 11

severe injury or vegetative state N = 4

Good outcome or mild injury

N = 218

Submersion victim status at hospital discharge

N = 233

Methods

Study Setting and Population

This is a retrospective study of unintentional submersion incidents that occurred between January 2003 and December 2009 among children less than 18 years of age who were treated at an urban tertiary-care pediatric hospital in Houston, Texas. The hospital has an annual Emergency Department (ED) census of 80,000 patients per year with less than 1% of the visits related to unintentional submersions. Submersion was defined as an event in which the whole body, including the airway, was under water based on recommended Utstein style guidelines [20]. Prior studies have defined vomiting as spontaneous, ac- tive abdominal muscle contraction resulting in expulsion of stomach contents [18]. Children who were pronounced dead in the field, for whom outcome data at hospital discharge was not documented and children described as having passive release of water from the victim’s mouth during resuscitation or minor spitting-up were excluded.

Most submersion victims in the region are directly transported to one of two local pediatric tertiary-care hospitals. Some patients are first transported by Emergency Medical Services (EMS) to community hospitals from where they may be either discharged to home or subse- quently transferred to our hospital. EMS transport data and community hospital treatment records are available for these Transferred patients. In addition, there are victims who are not transported by EMS if their condition does not warrant (non-transports) and a few vic- tims who are declared dead at the scene and are transported to the morgue.

The outcomes studied were survival at 24 hours (alive or dead) and condition at hospital discharge. The latter was described as favorable (good, mild/moderate impairment) or poor (severe disability, vegeta- tive or dead). This was based on the victim’s neurologic condition at hospital discharge as determined by the treating physician: Favorable (Good: conscious, alert child at discharge with return to previous level of functioning or Mild: minor neurological deficit and able to function at previous level); Poor (Death or Vegetative state and dependent on others for care or no higher mental functioning). The outcomes were determined from the medical records after reviewing the physical ex- amination findings and the condition at discharge. Death was based on county fatality reports.

Data Sources

Data were obtained from hospital medical records, EMS records, and county fatality reports. Records with a primary or secondary Interna- tional Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis code for drowning injury (994.1) were reviewed. Only the first hospital visit for submersion victims who had multiple visits for submersion-related complications was selected. Data from medical charts were entered into a standardized data collection instru- ment [21] by a single person. Pre-hospital reports and fatality data were entered into the same data collection instrument. Unavailable data were labeled missing.

Data included age, gender, race/ethnicity, insurance (private, non- private, self-pay), co-morbidities which includED asthma, Type 1 diabe- tes, epilepsy, autism, developmental delay, intellectual disability, attention-deficit/hyperactivity disorder, history of prior head trauma, hemiparesis, and history of prematurity (yes/no), injury day (weekday: Monday- Friday; weekend Saturday and Sunday) and season (summer, non-summer), location of submersion (swimming pool, bathtub, free flowing water, still water, ocean, other), submersion time (b 5 minutes,

>=5 minutes), supervision (parent, non-parent, unsupervised), mode of resuscitation (cardiopulmonary resuscitation (CPR), rescue breaths, chest compressions, back thrusts, other, none), apnea/intubation in ED, Glasgow Coma Score at ED presentation, hospitalization (yes/no), status at hospital discharge (good/mild injury, severe injury/ vegetative/dead), emesis (yes/no), and timing of emesis (pre-resuscitation, during/post-resuscitation). The study was approved by the institutional review board of Baylor College of Medicine.

Data verification

Duplicate records were discarded. In order to remove duplicate re- cords between hospital, fatality and EMS data; we matched cases by the age and date of occurrence of the injury event. Then we matched cases by name, gender, address of the injury and date of birth. Data were then transferred into an electronic database and reviewed by one investigator. (KF)

Statistical Analysis

Descriptive comparisons between emesis groups (yes/no) and cate- gorical covariates were analyzed using Pearson Chi-Square and Mann- Whitney tests for non-normalized distributed continuous factors. Based

Table 1

Descriptive comparison of pediatric submersion victims based on occurrence of emesis (N = 246)

Table 2 Resuscitation characteristics of pediatric submersion victims based on occurrence of eme- sis (N = 246)

	Emesis absent	Emesis present	All cases	p-Value			No Emesis	Yes Emesis	All (N = 246;	p-Value
	(n = 101;	(n = 145;	(N = 246;				(n = 101;	(n = 145;	100.0%) N (%)
	41.1%) N (%)	58.9%) N (%)	100.0%) N (%)				41.1%) N (%)	58.9%) N (%)	or Median
	or Median	or Median	or Median				or Median	or Median	(IQR)
	(IQR)	(IQR)	(IQR)				(IQR)	(IQR)
Age	3.0 (1.0, 5.0)	3.0 (1.50, 5.0)	3.0 (1.0, 5.0)	0.43		Resuscitation				0.25
Missing 0		0	0		CPR		55 (72.4)	74 (59.7)	129 (64.5)
Gender				0.80	Rescue		13 (17.1)	26 (21.0)	39 (19.5)
Male	66 (65.3)	97 (66.9)	163 (66.3)		Breaths
Female	35 (34.7)	48 (33.1)	83 (33.7)		Other?		8 (10.5)	23 (18.5)	31 (15.5)
Missing	0	0	0		None		0 (0.0)	1 (0.8)	1 (0.5)
Race/Ethnicity				0.60	Missing		25	21	46
Caucasian	36 (37.1)	63 (44.1)	99 (41.3) Patient 0.52
Hispanic	24 (24.7)	36 (25.2)	60 (25.0)		Hospitalized
African American	31 (32.0)	35 (24.5)	66 (27.5)		No		15 (14.9)	26 (17.9)	41 (16.7)
Other	6 (6.2)	9 (6.3)	15 (6.3)		Yes		86 (85.1)	119 (82.1)	205 (83.3)
Missing Injury Day	4	2	6	0.80	Missing Submersion		0	0	0	b0.001a
Weekday	56 (55.4)	78 (53.8)	134 (54.5)		Type
Weekend	45 (44.6)	67 (46.2)	112 (45.5)		Non-fatal		90 (89.1)	145 (100.0)	235 (95.5)
Missing	0	0	0		Fatal 11 (10.9) 0 (0.0) 11 (4.5)
Injury Season				0.99	Missing 0 0 0
Non-Summer	14 (13.9)	20 (13.8)	34 (13.8) Apnea or 0.01
Summer	87 (86.1)	125 (86.2)	212 (86.2) Intubated in
Missing	0	0	0 ED
Submersion Location				0.26	No		81 (83.5)	134 (93.7)	215 (89.6)
Swimming pool	69 (69.7)	118 (82.5)	187 (77.3)		Yes		16 (16.5)	9 (6.3)	25 (10.4)
Bathtub	21 (21.2)	17 (11.9)	38 (15.7) Missing 4 2 6
Free flowing water	1 (1.0)	2 (1.4)	3 (1.2) Timing of Emesis -
Still water	4 (4.0)	4 (2.8)	8 (3.3)		Pre-		0	20 (13.8)	20 (13.8)
Ocean	2 (2.0)	1 (0.7)	3 (1.2)		resuscitation
Other	2 (2.0)	1 (0.7)	3 (1.2)		Post-		0	125 (86.2)	125 (86.2)
Missing	2	2	4		resuscitation
Supervision				0.33	Missing		0	0	0

ED GCS 15.0 (10.0, 15.0) 15.0 (15.0, 15.0) 15.0 (15.0, 15.0) b0.001

None	4 (4.6)	12 (9.4)	16 (7.4)
Non-Parent	32 (36.8)	39 (30.5)	71 (33.0)
Parent	51 (58.6)	77 (60.2)	128 (59.5)
Missing	14	17	31
Submersion duration			0.89
b5 minutes	46 (85.2)	68 (86.1)	114 (85.7)
>=5 minutes	8 (14.8)	11 (13.9)	19 (14.3)
Missing	47	66	113
Comorbid conditions No	66 (77.6)	101 (79.5)	0.74 167 (78.8)
Yes	19 (22.4)	26 (20.5)	45 (21.2)
Missing	16	18	34
Insurance				0.84
Public	36 (38.7)	52 (36.6)	88 (37.4)
Private	25 (26.9)	44 (31.0)	69 (29.4)
Self-Pay	14 (15.1)	17 (12.0)	31 (13.2)
Other	18 (19.4)	29 (20.4)	47 (20.0)
Missing	8	3	11

Missing 30 23 53

GCS: Glasgow Coma Score. ED: Emergency Department.

^a P-value calculated using Fisher’s Exact Test.

* Chest compressions and back thrusts.

^a p-value calculated using Fisher’s Exact Test.

Comorbid conditions include asthma, Type 1 diabetes, epilepsy, autism, developmental delay, intellectual disability, attention-deficit/hyperactivity disorder, history of prior head trauma, hemiparesis, and history of prematurity.

on cross-tabulation cell frequencies between vomiting and the two out- come variables, 24-hour mortality and status at discharge, the Fisher’s Exact Test was utilized when cell values were less than five to assess sig- nificant differences between groups. All statistics were conducted using SPSS 22.0 statistical software (IBM Corporation, Somers, USA).

Results

There were 281 submersion victims with 246 patients (88%) in whom the presence or absence of emesis at the time of rescue or resus- citation was documented (Fig. 1). Submersion victims who did not have any emesis documented in their records (n = 35) did not differ in age, gender, race, submersion location or submersion duration and existing medical conditions from those in whom it was documented. They

were more likely to be publically insured, had fewer fatalities and more hospitalizations (data not shown).

For submersion victims in whom the presence or absence of emesis was documented, the median age was 3 years; 66% were males. Most in- cidents occurred at swimming pools (77%) and bathtubs (16%). Major- ity were hospitalized (83%). There were no differences in the age, gender, socio-demographic characteristics, submersion location and du- ration, resuscitation method or Need for hospitalization (Table 1). How- ever, those who had emesis around the time of rescue/resuscitation were significantly less likely to have apnea or be intubated in the ED, have a low ED GCS or die (Table 2). When evaluating 24-hour mortality and status at hospital discharge, no patient who had emesis died at 24 hours or had a poor outcome (Table 3) with the results being highly sig- nificant. Sub-analysis of victims based on the occurrence of emesis be- fore or after rescue/resuscitation (Table 4) showed that there were no differences in age, gender, socio-demographic characteristics, submer- sion location and duration, intubation or apnea in the ED or low ED GCS. However, victims who had emesis post rescue/resuscitation were significantly more likely to receive CPR or chest compressions than res- cue breaths.

Discussion

We observed that the presence of emesis at the time of resuscitation in childhood victims of submersions was associated with favorable out- comes at 24 hours post-submersion and at hospital discharge. This is con- trary to the literature that describes emesis as a negative prognostic indicator in victims of submersions [18]. Emesis is believed to threaten

Table 3

Relation of emesis anytime and outcome in pediatric submersions

Death at 24 hours N (%)

Yes No p-value^a Emesis absent 6 (100.0) 95 (39.6) 0.004

Emesis present 0 (0.0) 145 (60.4)

Table 4 (continued)

Emesis Pre- Resuscitation N= 20 (13.8%) N (%)

or Median (IQR)

Emesis Post- Resuscitation N = 125

(86.2%) N (%)

or Median (IQR)

p-Value

Outcome at hospital discharge

Public 5 (25.0) 47 (38.5)

Private 7 (35.0) 37 (30.3)

	Poor	Good	p-valuea	Self-Pay	5 (25.0)	12 (9.8)
Emesis absent	15 (100.0)	81 (37.2)	b 0.001	Other	3 (15.0)	26 (21.3)

Emesis present 0 (0.0) 137 (62.8)

^a Fisher’s Exact Test - Two-sided p-value.

Table 4

Sub-analysis of pediatric submersions comparing time of emesis and resuscitation (n = 145)

	Emesis Pre- Resuscitation	Emesis Post- Resuscitation	p-Value
	N = 20	N = 125
	(13.8%) N (%)	(86.2%) N (%)
	or Median	or Median
	(IQR)	(IQR)
Age	3.0 (2.0, 4.75)	3.0 (1.0, 5.0)	0.87
Missing Gender	0	0	0.48
Male	12 (60.0)	85 (68.0)
Female	8 (40.0)	40 (32.0)
Missing Race/Ethnicity	0	0	0.77
Caucasian	8 (40.0)	55 (44.7)
Hispanic	4 (20.0)	32 (26.0)
African American	6 (30.0)	29 (23.6)
Other	2 (10.0)	7 (5.7)
Missing	0	2
Injury Day			0.91
Weekday	11 (55.0)	67 (53.6)
Weekend	9 (45.0)	58 (46.4)
Missing	0	0
Injury Season			0.31a
Non-Summer	1 (5.0)	19 (15.2)
Summer	19 (95.0)	106 (84.8)
Missing	0	0
Submersion Location			0.19
Swimming pool	16 (80.0)	102 (82.9)
Bathtub	3 (15.0)	14 (11.4)
Free flowing water	0 (0.0)	2 (1.6)
Still water	0 (0.0)	4 (3.3)
Ocean	0 (0.0)	1 (0.8)
Other	1 (5.0)	0 (0.0)
Missing	0	2
Supervision			0.33
None	0 (0.0)	12 (10.8)
Non-Parent	5 (29.4)	34 (30.6)
Parent	12 (70.6)	65 (58.6)
Missing	3	14
Submersion duration			1.00a
b5 minutes	7 (87.5)	61 (85.9)
>=5 minutes	1 (12.5)	10 (14.1)
Missing	12	54
Resuscitation			0.001
CPR	2 (25.0)	72 (62.1)
Rescue Breaths	3 (37.5)	23 (19.8)
Other	2 (25.0)	21 (18.1)
None	1 (12.5)	0 (0.0)
Missing	12	9

Comorbidities 0.19

No 13 (68.4) 88 (81.5)

Yes 6 (31.6) 20 (18.5)

Missing 1 17

Patient Hospitalized b0.001

No	10 (50.0)	16 (12.8)
Yes	10 (50.0)	109 (87.2)

Missing 0 0

Insurance 0.20

(continued on next page)

Missing 0 3

Apnea or Intubated in ED 0.36^a

No 20 (100.0) 114 (92.7)

Yes 0 (0.0) 9 (7.3)

Missing 0 2

ED GCS 15.0 (14.75, 15.0) 15.0 (15.0, 15.0) 0.37

Missing 6 17

ED: Emergency department. GCS: Glasgow Coma Score.

Comorbid conditions include asthma, Type 1 diabetes, epilepsy, autism, developmental delay, intellectual disability, attention-deficit/hyperactivity disorder, history of prior head trauma, hemiparesis, and history of prematurity.

^a p-value calculated using Fisher’s Exact Test.

efforts to maintain a patent airway and increases the risk of complications secondary to aspiration, but vomiting is also an expected consequence of cardiopulmonary resuscitation maneuvers [22]. In performing chest com- pressions and abdominal thrusts, bystanders may apply force over the stomach and stimulate vomiting. Similarly, excessive rescue breaths may force external air into the stomach, increase gastric pressures and produce reflex emesis.

We believe that in pediatric submersions, vomiting may be a surro- gate marker for successful resuscitation. The fact that 86% of our patients who experienced emesis did so either during or after resuscitation sug- gests that resuscitation was a likely precipitating factor for emesis. A sub-analysis of submersion victims who had emesis revealed that eme- sis is more likely to occur with CPR or chest compression than rescue breaths. We were unable to explain the cause of emesis in a sixth of our patients who experienced emesis prior to any rescue interventions. We speculate that when vomiting occurs without external provocation, patient may have sufficient neurologic capacity to identify the presence of unwanted gastric contents and produce the necessary muscular con- tractions for their expulsion. Alternatively, severe nausea and vomiting can result from respiratory or metabolic acidosis, both of which occur with the hypoxemia of prolonged submersion [23].

Undoubtedly, all victims of Submersion injury who experience vomiting should at some point be evaluated for acid-base disturbances and possible aspiration pneumonitis; however, it is worth noting that the absence of aspiration with emesis does not preclude the child from other forms of lung damage that occur with submersion injuries. Water aspiration can produce significant pulmonary edema resulting ei- ther from Fluid shifts dependent on the tonicity of the water or from sur- factant wash-out with subsequent alveolar damage [24]. The American Heart Association explains that hypoxia with prolonged water submer- sion may also increase capillary permeability and produce pulmonary edema [22]. Submersion is, therefore, an independent risk factor for acute respiratory distress syndrome secondary to edema with or with- out aspiration of vomitus [1].

The benefits of prompt cardiopulmonary resuscitation efforts, which work to correct this hypoxemia, are clear [13] and should be employed regardless of possible secondary complications. However, this study suggests that the risk of vomiting may be overstated and that emesis may not be a significant cause of death or severe morbidity following submersion injury. Since the placement of nasogastric (NG) tubes for gastric suctioning is not a benign process, perhaps more cautious con- sideration should be given to NG tube application for the purpose of preventing emesis. Further studies to evaluate the risks and benefits of

NG tube placement in victims of submersion injury are necessary to pro- vide a more definitive recommendation.

Finally, though this study indicates that emesis may be protective for submersion injury, this appears to be largely due to the benefits of car- diopulmonary resuscitation. Early rescue and good quality CPR are par- amount in avoiding poor outcomes following submersions. We do not recommend routine induction of emesis in the ED. Rather, providers should focus on continued cardiopulmonary care, including ensuring adequate Ventilation and oxygenation, and correction of respiratory and metabolic acidosis by conventional means.

There were limitations in this study. First, we were unable to obtain data on all submersion events in Harris County and surrounding areas; therefore, the results underestimate the true burden of submersion in- cidents. These would include non-life-threatening submersions for which EMS was not called, victims treated at outpatient physician of- fices, or children who did not visit the hospital emergency department. Second, missing data were another handicap in the study due to the ret- rospective nature of the study. Death certificates and ED records are known to have incomplete information on the victim’s activities, swim- ming ability, prior health conditions, use of protective equipment and rescue personnel, supervision and cardiopulmonary resuscitation.¹ Fi- nally, since this is a single center study our results may not be general- izable to other areas of the country. Further studies will need to be conducted to confirm our results.

Conclusions

Emesis in pediatric submersion victims is inversely associated with death at 24 hours or poor outcomes at hospital discharge. The relation- ship between emesis and the adequacy of resuscitation of pediatric sub- mersion victims needs to be further studied.

Conflict of interest

The authors do not have any conflicts of interest to report.

Acknowledgment

The study did not receive grant support. The authors would like to thank Dr. Andrea T Cruz for helpful advice regarding the article.

References

1. Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report. Drowning - United States, 2005-2009. MMWR Morb Mortal Wkly Rep 2012;61(19): 344-7.
2. Bowman SM, Aitken ME, Robbins JM, Baker SP. Trends in US pediatric drowning hos-

   pitalizations, 1993-2008. Pediatrics 2012;129:275-81.

   Brenner RA, Tremble AC, Smith GS, Kessler EP, Overpeck MD. Where children drown, United States, 1995. Pediatrics 2001;108:85-9.

3. Centers for Disease Control and Prevention, National Center for Health Statistics. Compressed Mortality File 1999-2012 on CDC WONDER Online Database, released October 2014. Data are from the Compressed Mortality File 1999-2012 Series 20 No. 2R. Accessed at http://wonder.cdc.gov/cmf-icd10.html; 2014. on January 5, 2015.
4. Blum C, Shield J. Toddler drowning in domestic swimming-pools. Inj Prev 2000;6:288-90.
5. Pelletier AR, Gilchrist J. Fatalities in swimming-pools with lifeguards: USA, 2000-2008. Inj Prev 2011;17:250-3.
6. Quan L, Cummings P. Characteristics of drowning by different age groups. Inj Prev 2003;9:163-8.
7. Saluja G, Brenner RA, Trumble AC, Smith GS, Schroeder T, Cos C. Swimming-pool drownings among US residents aged 5-24 years: understanding racial/ethnic dis- parities. Am J Public Health 2006;96:728-73.
8. Szpilman D, Bierens JJLM, Handley AJ, et al. Drowning. N Engl J Med 2012;366:2102-10.
9. Weiss J. Committee on Injury, Violence and Poison Prevention. Prevention of drown- ing. Pediatrics 2010;126:e253-62.
10. Habib DM, Tecklenburg FW, Webb SA, Anas NG, Perkin RM. Prediction of childhood drowning and near-drowning morbidity and mortality. Pediatr Emerg Care 1996; 12(4):255-8.
11. Quan L, Kinder D. Pediatric submersions: prehospital predictors of outcome. Pediat- rics 2001;90(6):909-13.
12. Quan L, Wentz KR, Gore EJ, Copass MK. Outcome and predictors of outcome in pedi- atric submersion victims receiving prehospital care in King County, Washington. Pe- diatrics 1990;86(4):586-93.
13. Laughlin JJ, Eigen H. pulmonary function abnormalities in survivors of near drown- ing. J Pediatr 1982;100(1):26-30.
14. Mackie IJ. Patterns of drowning in Australia, 1992-1997. Med J Aust 1999; 171(11-12):587-90.
15. Manolios N, Mackie I. Drowning and near-drowning on Australian beaches patrolled by savers: a 10-year study, 1973-1983. Med J Aust 1988;148(4):165-7 [170-1].
16. Orlowski JP, Szpilman D. Drowning: rescue, resuscitation, and reanimation. Pediatr Clin N Am 2001;48(3):627-46.
17. Fenner PJ, Harrison SL, Williamson JA, Williamson BD. Success of surf lifesaving re- suscitations in Queensland 1973-1992. Med J Aust 1995;163:580-3.
18. Macnab AJ. Effective interventions for nearly drowned children. Can Fam Physician 1995;41:1545-56.
19. Idris AH, Berg RA, Bierens J, et al. Recommended guidelines for uniform reporting of data from drowning: the “Utstein Style.”. Circulation 2003;108:2565-74.
20. Texas submersion reporting form. Available at http://www.houstontx.gov/health/ Epidemiology/subform.pdf. [accessed on May 6, 2012].
21. American Heart Association. Part 8: advanced challenges in resuscitation: section 3: special challenges in ECC. Circulation 2000;102. http://dx.doi.org/10.1161/01.CIR. 102.suppl_1.I-299 [I-299-I-252].
22. Segarra F, Redding RA. Modern concepts about drowning. Can Med Assoc J 1974; 110(9):1057,1059-60.
23. Fleetham JA, Munt PW. Near-drowning in Canadian waters. Can Med Assoc J 1978; 118:914-7.