Hypothermia is defined as a body temperature less than 35°C and is subdivided into mild (32°–35°C), moderate (28°–32°C), and severe (<28°C). Primary hypothermia also known as ‘accidental hypothermia’ is due to environmental exposure causing disruption of thermoregulation, and secondary hypothermia results from various underlying medical illnesses unrelated to cold exposure. The Center for Disease Control and Prevention reported a total of 16 911 deaths in the USA due to exposure to excessive natural cold with an average of 1301 per year from 1999 to 2011 and ∼67% of these hypothermia-related deaths among males 1. Hypothermia has been associated with increased mortality in those presenting with trauma and severe sepsis 2, and is considered an important prognostic tool. Hypothermia affects all body organs but consequences on the cardiovascular system are most serious because severe hypothermia is a precursor for ventricular fibrillation. Although the outcome of severely hypothermic patients may be unfavorable, early recognition and treatment can simply reverse this process and improve outcome.
Hypothermia can present with misleading symptoms, such as confusion, dizziness, chills, or dyspnea leading to delayed diagnosis. Moreover, the diagnosis of hypothermia can be initially missed by paramedics with the commonly used oral or tympanic thermometers since standard oral thermometers record temperature only as low as 35°C accurately. Paramedics also rarely use thermometers in evaluating patients in the prehospital setting and ECG features of hypothermia may therefore be of value. Detecting hypothermia in the field is important as it will allow for early management which involve preventing further heat loss, gradual rewarming, and more importantly watching for cardiac dysrhythmias which begin to develop with core body temperature less than 30°C. Emergency and intensive care physicians are then the next responders to hypothermic subjects with environmental exposure, sepsis, and trauma and should be aware of the ECG features of hypothermia.
The effect of hypothermia on the ECG is not just limited to the classical emergence of the Osborn wave, but also includes effects on the intervals leading to prolongation of PR, QRS and QT intervals, conduction abnormalities, and the propensity to cause various atrial and ventricular arrhythmias. The magnitude of these effects relate to the degree of hypothermia. ECG findings in hypothermia may also simulate other disease entities (e.g. acute coronary syndrome). Awareness that hypothermia can masquerade as myocardial ischemia would avert the unnecessary administration of potentially harmful medications like anticoagulant or thrombolytic therapy in subjects who may be coagulopathic from hypothermia 3. The effect of hypothermia on the ECG should be known to various healthcare providers expected to encounter these cases and not just cardiologists 4. Because of the important clinical and prognostic implications of hypothermic ECG changes, we summarized the effect of various stages of hypothermia on the surface ECG as follows.
If hypothermia develops rapidly, it will lead to sympathoadrenal activation which will raise the plasma norepinephrine level thereby contributing to the initial sinus tachycardia evident in hypothermic subjects. Atrial fibrillation (AF) is a common arrhythmia observed in hypothermic patients occurring in 19–21% of the cases 5,6. Okada 7 analyzed the ECG of 60 patients with accidental hypothermia and found that AF occurred with moderate and severe hypothermia. In these patients, AF usually converted to sinus rhythm spontaneously when normothermia was restored without the need for antiarrhythmic agents 7. Higuchi et al.8 found that the presence of Osborn waves in hypothermic patients was associated with lower frequency of sinus rhythm (P=0.047). AF may be falsely over diagnosed in hypothermic patients due to shivering artifacts 6 evident in 56% of hypothermic cases 9 and the occasional inability to discern the P wave despite the preservation of the sinus rhythm 10. The effect of the development of AF on mortality in hypothermic patients has been controversial. While Okada 7 and Rankin and Rae 11 reported no significant difference in mortality between the AF and non-AF groups, Graham et al.5 noted that AF was associated with high mortality. However, these three studies comprised small samples and univariate analysis was likely the utilized statistical technique for comparison. Other tachyarrhythmias that can be encountered in hypothermic patients are ventricular arrhythmias which are the most serious complication one can expect in a hypothermic patient. There are few case reports demonstrating that the development of the Osborn wave in hypothermic patients may be a precursor for malignant ventricular arrhythmias. Although Osborn waves were found to be precursors for ventricular fibrillation in acute coronary syndrome and is associated with higher in-hospital mortality 12, this arrythmogenic potential of the Osborn wave in hypothermic subjects has not been confirmed 13.
Bradycardia is common in hypothermic patients 14. De Souza et al.15 found a significant correlation between the core body temperature and the heart rate (r=0.456, P=0.001), therefore the lower the body temperature, the more intense the bradycardia. Sinus bradycardia is the commonest bradyarrythmia in hypothermic subjects 16. De Souza et al.15 found the incidence of sinus bradycardia to be 52.5% of patients admitted with accidental hypothermia. Other bradyarrythmias are due to the blocking effect of hypothermia on the atrioventricular (AV) node leading to junctional and idioventricular rhythms 9. The incidence of junctional, idioventricular rhythm and AV block was reported to be 16.9% in the study by De Souza et al.15. Graham et al.5 found a 5% incidence of junctional bradycardia. Nonetheless, junctional bradycardia may be over diagnosed in hypothermic patients due to reduction of the amplitude of the P wave 9. Somerville 17 studied the effect of cooling on patients with AF receiving cardiac surgery under hypothermia and noticed slowing of the atrial and ventricular rate with every 1–3°C drops in body temperature. Higuchi et al.8 found that patients with Osborn waves had lower heart rate compared with those without Osborn waves (P=0.002).
Decrease in amplitude of the P wave
A decrease in the amplitude of the P wave occurs with progression of hypothermia 7. This may lead to the false interpretation of the ECG as junctional rhythm despite maintenance of the native sinus rhythm. Zotti et al.10 studied the effect of hypothermia on the conduction system of the dog’s heart, and noted that sinus impulse transmission to the atrial septum can occur without an evident P wave on the ECG. This was also confirmed by Jacob et al.18 who reported a case of accidental hypothermia with absent P waves that was found to have – on His bundle ECG – sinoventricular conduction and a prolonged AH interval not responsive to atropine. These abnormalities were reversed after rewarming 18. The inability to discern the P wave in hypothermic subjects may also lead to false interpretation of the ECG as AF despite preservation of the sinus rhythm 10.
Prolongation of the PR, ORS and QT intervals
It was found that prolonged or moderate hypothermia can lead to first degree AV block 9,19,20 which resolves after rewarming therapy. Worsening hypothermia also leads to increased QRS duration due to prolongation of ventricular depolarization 9,19,20. As core body temperature decreases, the QTc interval also prolongs which was found to occur in about 73% of hypothermia patients due to delayed ventricular repolarization 15. The prolongation of PR, QRS, and QT interval with reduction of body temperature was exemplified in many case reports 19. Higuchi et al.8 observed that patients with hypothermic Osborn waves, had longer PR interval (P=0.013), QRS interval (P<0.0001), and QT interval (P=0.0011).
The emergence of the Osborn waves
The J point in ECG is the point where the QRS complex joins the ST segment and represents the end of depolarization and the beginning of repolarization. The J wave is produced when the J point is markedly deviated from the baseline. This wave has received a variety of names but the ‘Osborn wave’ and the ‘J wave’ are interchangeably the most commonly used terms when referring to this wave. We know that the development of Osborn waves in the ECG is not pathognomonic of hypothermia (some cases have non-hypothermic Osborn waves) and that hypothermic patients with Osborn waves have lower core body temperature compared with those without Osborn waves (P<0.001) 8. However the opposite is not necessarily true; one should not expect Osborn waves to emerge in all hypothermic patients. Studies on accidental hypothermia reported variable incidence of development of the Osborn wave. While some studies showed no correlation 14,21 between the depth of hypothermia and the appearance of Osborn waves, others showed a correlation between lower temperature and emergence of the Osborn wave 22,23. In studies by Okada et al.24 and Higuchi et al.8, Osborn waves were evident in 80% and in 50% of the cases who presented with accidental hypothermia, respectively. In the later study the frequency of appearance of the Osborn wave was 100% with severe hypothermia, 75% with moderate hypothermia and 10.7% with mild hypothermia 8.
Hypothermia is likely not the sole determinant of the appearance of Osborn waves. This has been initially stated by Osborn 25 in his work on experimental dogs when he found that the appearance of the Osborn wave was related to acidosis as they were absent when the pH was normally maintained by mechanical ventilation. Edelman and Joynt 26 later illustrated this observation in a case where the Osborn waves disappeared with correction of pH from 7.03 to 7.33 despite the same body temperature of 92°F. Further studies are needed to illustrate the predictors for the development of Osborn waves in hypothermia.
The amplitude of the Osborn wave increases as the body temperature decreases
Multiple case reports illustrated the inverse correlation between the amplitude of the Osborn wave and core body temperature (Fig. 1) 19. Omar and Camporesi 27 analyzed the ECG (admission as well as follow-up ECG during rewarming) of all published case reports with hypothermic Osborn waves and found a strong inverse correlation (r=−0.682, P<0.001) between the Osborn wave amplitude and core body temperature when all admission and follow-up ECGs were included therefore one should expect a reduction in the amplitude of the Osborn wave with rewarming and an increase in the amplitude with further decline of body temperature. Nonetheless, when analyzing admission-only ECG of all reported cases of hypothermic Osborn waves, this correlation was only moderate (r=−0.410, P<0.001) 27. The stronger correlation between the Osborn wave and core body temperature in the group of cases with more than one recorded ECG (during hospitalization) compared with analyzing admission-only ECG suggest that the body temperature is not the sole determinant of the amplitude of the Osborn wave and that other factors likely contribute to its development. The effect of these factors is minimized when looking at all ECGs (admission and follow-up ECGs), and is exaggerated when looking only at a single initial ECG of different cases. We therefore concluded that the amplitude of the Osborn wave in the admission ECG of hypothermic patients may not accurately predicts the core body temperature which provokes the immediate rectal temperature measurement whenever Osborn waves are evident. In most published case reports, the amplitude of the Osborn wave gradually decreased during hospitalization with rewarming and correction of the underlying primary cause for hypothermia. However, in few cases when hypothermia was due to an irreversible process, such as brainstem death, where the core body temperature continued to decrease with simultaneous increase in the amplitude of the Osborn wave 28.
The Osborn waves become evident in more ECG leads and not just the inferolateral leads
Yan et al. experimentally reproduced the Osborn waves in a canine wedge-shaped preparation and the waves were noticed more in the inferior and lateal leads 29. The Osborn wave mean vector axis was directed towards the left ventricle and septum and he concluded that the Osborn wave in the dog is most prominent in these leads. Moreover, vectorcardiography indicates that the Osborn wave forms an extra loop that occurs at the junction of the QRS and T loops and is directed leftward and anteriorly, which explains its prominence in leads associated with the left ventricle 30. This has been confirmed in many of the published case reports of hypothermic Osborn waves that demonstrated its emergence in the inferolateral leads. However, with progression of hypothermia, the Osborn waves tend to predominate in all ECG leads and not just the inferolateral leads 19. In an observational study of 60 consecutive patients admitted with accidental hypothermia, Higuchi et al.8 found that Osborn waves were evident in the inferolateral leads in 70% of the cases. Moreover, the number of sites where Osborn waves were evident was significantly higher in the severe hypothermia group compared with the mild and moderate hypothermia groups (P<0.0001) 8.
Shivering artifacts emerge
Shivering artifacts are common in hypothermic patients and can be mistaken for AF 5. It was found to occur in 56% of patient with hypothermia 5. A higher proportion of survivors had shivering artifacts in their ECG (66 vs. 38% of survivors and nonsurvivors, respectively, P=0.047) suggesting that the inability to mount a shivering response is associated with poor outcome 5. One exception is the head trauma victim who develops brainstem death with subsequent hypothermia due to of impaired thermoregulatory ability from hypothalamic dysfunction; in this instance hypothermia is not associated with shivering artifacts 28. We have previously reported cases of brain death-induced hypothermia that were not on any sedation and had Osborn waves without shivering artifacts in the surface ECG 19,28. We hypothesized that, in head trauma victims, specifically those who are not on any sedation, the development of hypothermic Osborn waves without shivering artifacts should serve as a clue for suspecting brain death. Its presence should call for assessment of the patient looking for reliable signs to confirm or refute the diagnosis of brain death. Notice in Fig. 2, the occurrence of high amplitude Osborn waves in a hypothermic patient (26.8°C) due to traumatic brainstem death but with absence of shivering artifacts.
Conflicts of interest
There are no conflicts of interest.
2. Kushimoto S, Yamanouchi S, Endo T, Sato T, Nomura R, Fujita M, et al.. Body temperature abnormalities in non-neurological critically ill patients: a review of the literature. J Intensive Care 2014; 2:14.
3. Glusman A, Hasan K, Roguin N. Contraindication to thrombolytic therapy in accidental hypothermia simulating acute myocardial infarction. Int J Cardiol 1990; 28:269–272.
4. Omar HR, Helal E, Camporesi EM. The Osborn wave, not just for cardiologists. Cardiovasc Endocrinol 2012; 1:33–34.
5. Graham CA, McNaughton GW, Wyatt JP. The electrocardiogram in hypothermia. Wilderness Environ Med 2001; 12:232–235.
6. Vassallo SU, Delaney KA, Hoffman RS, Slater W, Goldfrank LR. A prospective evaluation of the electrocardiographic manifestations of hypothermia. Acad Emerg Med 1999; 6:1121–1126.
7. Okada M. The cardiac rhythm in accidental hypothermia. J Electrocardiol 1984; 17:123–128.
8. Higuchi S, Takahashi T, Kabeya Y, Hasegawa T, Nakagawa S, Mitamura H. J waves in accidental hypothermia. Circ J 2014; 78:128–134.
9. Mattu A, Brady WJ, Perron AD. Electrocardiographic manifestations of hypothermia. Am J Emerg Med 2002; 20:314–326.
10. Zotti EF, Lesage MA, Wallace AG, Sealy WC, Young WG Jr. Conduction and excitability of the intact hypothermic heart studies using chronically implanted electrodes and controlled deep hypothermia. Trans Am Soc Artif Intern Organs 1965; 11:139–142.
11. Rankin AC, Rae AP. Cardiac arrhythmias during rewarming of patients with accidental hypothermia. Br Med J (Clin Res Ed) 1984; 289:874–877.
12. Aïssou L, Hermida JS, Traullé S, Delaverhne A, Diouf M, Leborgne L, et al.. Prevalence and prognostic significance of ‘J waves’ in patients experiencing ventricular fibrillation during acute coronary syndrome. Arch Cardiovasc Dis 2012; 105:578–586.
13. Omar HR. The arrythmogenic potential of the Osborn wave. Am J Emerg Med 2013; 31:424–426.
14. Tolman KG, Cohen A. Accidental hypothermia. Can Med Assoc J 1970; 103:1357–1361.
15. De Souza D, Riera AR, Bombig MT, Francisco YA, Brollo L, Filho BL, et al.. Electrocardiographic changes by accidental hypothermia in an urban and a tropical region. J Electrocardiol 2007; 40:47–52.
16. Chhabra L, Devadoss R, Liti B, Spodick DH. Electrocardiographic changes in hypothermia: a review. Ther Hypothermia Temp Manag 2013; 3:54–62.
17. Somerville W. The effect of hypothermia on atrial fibrillation and other arrhythmias. Br Heart J 1960; 22:515–521.
18. Jacob AI, Lichstein E, Ulano SD, Chadda KD, Gupta PK, Werner BM. A-V block in accidental hypothermia. J Electrocardiol 1978; 11:399–402.
19. Omar HR, Abdelmalak HD. Osborn waves: an inverse correlation with core body temperature. Cleve Clin J Med 2011; 78:734–736.
20. Carden D, Doan L, Sweeney PJ, Nowak RM. Hypothermia. Ann Emerg Med 1982; 11:497–503.
21. Duguid H, Simpson RG, Stowers JM. Accidental hypothermia. Lancet 1961; 2:1213–1219.
22. Emslie-Smith D, Sladden GE, Stirling GR. The significance of changes in the electrocardiogram in hypothermia. Br Heart J 1959; 21:343–351.
23. Clements SD Jr, Hurst JW. Diagnostic value of electrocardiographic abnormalities observed in subjects accidentally exposed to cold. Am J Cardiol 1972; 29:729–734.
24. Okada M, Nishimura F, Yoshino H, Kimura M, Ogino T. The J wave in accidental hypothermia. J Electrocardiol 1983; 16:23–28.
25. Osborn JJ. Experimental hypothermia; respiratory and blood pH changes in relation to cardiac function. Am J Physiol 1953; 175:389–398.
26. Edelman ER, Joynt K. J waves of Osborn revisited. J Am Coll Cardiol 2010; 55:2287.
27. Omar HR, Camporesi EM. The correlation between the amplitude of Osborn wave and core body temperature. Eur Heart J Acute Cardiovasc Care 2014. [Epub ahead of print].
28. Omar HR, Rashad R, Helal E. The giant waves of Osborn in brain death. Intern Med J 2011; 41:841–842.
29. Yan GX, Antzelevitch C. Cellular basis for the electrocardiographic J wave. Circulation 1996; 93:372–379.
30. Blades B, Pierpont HC. A simple method for inducing hypothermia. Ann Surg 1954; 140:557–563.