Clinical Chemistry / Analytes in Collapsed Marathon Runners
Cardiac Biomarkers, Electrolytes, and Other Analytes
in Collapsed Marathon Runners
Implications for the Evaluation of Runners Following Competition
Arthur J. Siegel, MD,1,4 James Januzzi, MD,2,4 Patrick Sluss, PhD,3,4 Elizabeth Lee-Lewandrowski, MPH, PhD,3,4
Malissa Wood, MD,2,4 Terry Shirey, PhD,5 and Kent B. Lewandrowski, MD3,4
Key Words: Cardiac troponin T; B-type natriuretic peptide; N-terminal pro-B-type natriuretic peptide; Lactic acidosis; Ionized magnesium;
Acute coronary syndrome; Sudden cardiac death; Marathon
DOI: 10.1309/4L0M60MGAQBCHMV7
Abstract
We measured analytes in collapsed Boston
Marathon runners to compare with changes in
asymptomatic runners. Of collapsed runners at the
2007 marathon, 18.2% had a measurable cardiac
troponin T (cTnT) value with a mean postrace level
of 0.017 ng/mL (0.017 g/L; SD, 0.02 ng/mL [0.02
g/L]). Three subjects had cTnT values above the
cutoff (0.10 ng/mL [0.10 g/L]) typically used for the
diagnosis of acute myocardial infarction. The mean
and median N-terminal pro-B-type natriuretic peptide
levels were 73 ng/L (SD, 77.3 ng/L) and 54.3 ng/L
(interquartile range, 22.8-87.3 ng/L), respectively, in
collapsed runners. Only 4.9% had values more than
the age-specific normal value (<125 ng/L for subjects
younger than 75 years). In collapsed subjects at the
2006 marathon, 18.0% had an abnormal sodium value,
including 18 cases of hypernatremia and 7 cases of
hyponatremia. The ionized calcium level was low in
49% of subjects, and the ionized magnesium level was
low in 19.5% and elevated in 1 subject. The blood
lactate level was elevated in 95% of subjects.
The frequency of elevated postrace cTnT levels in
collapsed athletes after endurance exercise is similar
to that in asymptomatic runners. Other metabolic
abnormalities, including hypernatremia, hyponatremia,
low ionized calcium and magnesium levels, and lactic
acidosis may contribute to muscle fatigue and collapse.
948 Am J Clin Pathol 2008;129:948-951
948 DOI: 10.1309/4L0M60MGAQBCHMV7
Although regular physical activity reduces future coro-
nary heart disease events, episodes of vigorous exercise
such as marathon running may precipitate acute coronary
syndrome (ACS) and sudden death in 1 per 50,000 marathon
participants.1-5 Prior studies of changes in cardiac troponin
T (TnT) and I and B-type natriuretic peptide (BNP and
N-terminal pro-B-type natriuretic peptide [NT-proBNP])
levels in asymptomatic runners who completed a marathon
have demonstrated that mild to moderate elevations in levels
of these markers are common without apparent ACS or heart
failure.2,6-10 The mechanism of cardiac injury and myocyte
stress reflected by elevation of troponin and B-type natriuretic
peptide concentrations, respectively, remains under debate;
however, presumably these elevations result from an incom-
plete myocardial adaptation to training in which vulnerable
myocytes are selectively eliminated, particularly in athletes
with lesser training.11 As well, the effects of endurance exer-
cise on electrolyte and lactic acid concentrations as a function
of collapse following exercise remains unclear.
In this study, we evaluated the levels of TnT, NT-proBNP,
electrolytes, ionized calcium and magnesium, and lactate in
collapsed marathon runners in the medical tent at the finish
line of the 2006 and 2007 Boston marathons. Our hypothesis
was that more prevalent abnormalities in cardiac biomarker,
electrolyte, and lactic acid results would be noted in collapsed
runners compared with asymptomatic runners.
Materials and Methods
All study procedures were approved by the McLean
Hospital (Belmont, MA) Institutional Review Board.
Heparinized blood samples were obtained from collapsed
American Society for Clinical Pathology
 Clinical Chemistry / Original Article

marathon runners admitted to the finish line medical tent
at the 2006 and 2007 Boston marathons. Samples from the
2006 marathon were analyzed by whole blood point-of-care
technologies for electrolytes, glucose, serum urea nitrogen,
lactate, ionized calcium, and ionized magnesium using an
acute care analyzer (Nova Biomedical, Waltham, MA).
After the 2007 marathon, discarded plasma samples
were frozen at 80C for subsequent analysis of TnT and
NT-proBNP on the Roche Elecsys 2010 immunoassay ana-
lyzer (Roche Diagnostics, Indianapolis, IN) using a fourth-
generation TnT assay and the standard Roche NT-proBNP
assay. The cut point yielding the highest functional
sensitivity at a 10% coefficient of variation for the TnT
assay is 0.03 ng/mL (0.03 g/L), and the appropriate age-
matched upper limit of normal for the NT-proBNP assay
for this population is less than 125 ng/L. The imprecision
of the Roche NT-proBNP assay varies from less than 2.7%
(within-run) to less than 3.2% (total imprecision) across the
analytic range. Hemolyzed samples were excluded from
the study.
The mean age of the subjects was 39.1 years (SD, 10.4
years). Of the subjects, 62% were men and 38% were women.
Results
Mean values for whole blood point-of-care chemistry
analytes in collapsed marathon runners from the 2006 Boston
Marathon are shown in zTable 1z. Not all analytes were mea-
sured for each subject owing to the availability of analyzers
and samples for testing. Of 139 subjects, 25 (18.0%) had an
abnormal sodium value; 18 cases showed hypernatremia and
7 showed hyponatremia. The serum urea nitrogen level was
79 cases and whole blood glucose levels in 43 (31%) of 139
cases, with no hypoglycemia noted.
The mean TnT value was 0.017 ng/mL (0.017 g/L; SD,
0.02 ng/mL [0.02 g/L]) in 99 collapsed marathon runners
at the 2007 Boston Marathon zTable 2z. Of the 99 collapsed
runners, 18 (18%) exhibited measurable levels of TnT and 8
(8%) had values of more than the threshold of 0.03 ng/mL
(0.03 g/L) indicative of myocardial necrosis2 by a fourth-
generation TnT assay. Three subjects had TnT values of more
than the traditional cutoff point of 0.1 ng/mL (0.1 g/L) used
for the diagnosis of acute myocardial infarction (0.106, 0.172,
and 0.108 ng/mL [0.106, 0.172, and 0.108 g/L]).
The median NT-proBNP value was 54 ng/L (interquartile
range, 22.8-87.3 ng/L). Only 5 runners had NT-proBNP val-
ues of more than the age-adjusted cutoff value. These 5 values
showed modest elevations of NT-proBNP (186, 188, 224, 221,
and 331 ng/L). Of the 5 runners with an elevated NT-proBNP
level, 3 (60%) also had measurable levels of TnT.
Discussion
Multiple studies reporting elevations in cardiac TnT and
B-type natriuretic peptide (BNP or NT-proBNP) levels follow-
ing marathon running clearly establish a myocardial source, fur-
ther supported by recent data demonstrating strong associations
between elevated cardiac biomarkers and the postrace develop-
zTable 2z
Distribution of Troponin T Values in 99 Collapsed Marathon
Runners
Troponin T Range, ng/mL (g/L) No. (%) of Cases

elevated in 20 (49%) of 41. Of the subjects, 45 (51%) of 89 <0.01 (<0.01) 81 (82)

had a low ionized calcium level and 9 (22%) of 41 an abnor- >0.01 to <0.03 (>0.01 to <0.03) 10 (10)
mal ionized magnesium level that was low in 8 and elevated
0.03 to <0.10 (0.03 to <0.10) 5 (5)
>0.10 (>0.10) 3 (3)
in 1 sample. Blood lactate levels were elevated in 75 (95%) of

zTable 1z
Summary of Whole Blood Routine Chemistry Testing Results in Collapsed Marathon Runners*

Analyte No. of Samples Mean SD Minimum Maximum

Sodium (mmol/L) 139 142.3 4.1 129.6 152.2

Potassium (mmol/L) 139 4.31 0.62 3.41 8.22
Ionized calcium (mmol/L) 89 1.09 0.07 0.9 1.22
Hematocrit (%) 134 45 3.4 37 55
Glucose (mg/dL) 139 117 37 58 294
Lactate (mmol/L) 79 3.45 1.61 1.1 11.2
Urea nitrogen (mg/dL) 41 19.76 4.2 3 27
Ionized magnesium (mmol/L) 41 0.47 0.07 0.23 0.62
* Reference ranges are as follows: sodium, 135-147 mmol/L; potassium, 3.5-5 mmol/L; ionized calcium, 1.14-1.30 mmol/L; hematocrit: male, 41%-53%; female, 36%-46%;
glucose, 70-110 mg/dL; lactate, <2 mmol/L; urea nitrogen, 8-25 mg/dL; and ionized magnesium, 0.47-0.65 mmol/L. Values for sodium, potassium, ionized calcium, lactate, and
ionized magnesium are given in Systme International (SI) units; conversions to conventional units are as follows: sodium (mEq/L), divide by 1.0; potassium (mEq/L), divide
by 1.0; ionized calcium (mg/dL), divide by 0.25; lactate (mg/dL), divide by 0.111; ionized magnesium (mEq/L), divide by 0.50. Other values are given in conventional units;
conversions to SI units are as follows: hematocrit (proportion of 1.0), multiply by 0.01; glucose (mmol/L), multiply by 0.0555; urea nitrogen (mmol/L), multiply by 0.357.

American Society for Clinical Pathology Am J Clin Pathol 2008;129:948-951 949

949 DOI: 10.1309/4L0M60MGAQBCHMV7 949
Siegel et al / Analytes in Collapsed Marathon Runners
ment of abnormalities shown on 2-dimensional echocardiog-
raphy.9 In a recent study of asymptomatic marathon runners
following competition, more than 60% of participants had
measurable increases in TnT levels, with 40% at or above the
decision limit for acute myocardial necrosis (TnT 0.03 ng/
mL [0.03 g/L]). These changes were inversely related to train-
ing experience.9 In another recent study, postrace cardiac TnT
levels were elevated (0.01 or 0.1 ng/mL [0.01 or 0.1 g/L])
in 68% of 482 runners following the 2002 Boston Marathon,
including 11% with a TnT level of 0.075 ng/mL (0.075 g/L)
or more or of 0.5 ng/mL (0.5 g/L) or more.11
In contrast with these results in asymptomatic runners
who completed a marathon, in our study, only 18% of col-
lapsed runners showed measurable TnT values, with 8%
greater than 0.03 ng/mL (0.03 g/L); 3 had values of more
than 0.1 ng/mL (0.1 g/L). These percentages are distinctly
lower than those observed by Neilan et al9 and Fortescue
et al11 in asymptomatic runners. Because the degree of TnT
release seems to somewhat parallel the degree of endurance
exercise, it may be that our collapsed runners did not achieve
the intensive myocardial stress achieved by runners who com-
plete a full marathon. Lending further support to this concept
is that the values for NT-proBNP in collapsed runners in our
study were generally within the age-adjusted normal range
and lower than the median postrace NT-proBNP level of 131
pg/mL reported in asymptomatic runners who successfully
completed the race.9 Thus, in aggregate, our data suggest that
cardiovascular insufficiency does not underlie the mechanism
of collapse in the vast majority of collapsed runners.
Rapid whole blood point-of-care analyte results in 135
collapsed runners at the 2006 race demonstrated frequent
abnormalities of sodium, lactate, ionized calcium, and ionized
magnesium levels. Hypernatremia was observed in 13% and
hyponatremia in 5% of collapsed marathon runners (com-
pared with 25% and 6%, respectively, during the 2004 race
at warmer temperatures12). Prevention of hypernatremia and
hyponatremia is the focus of recently revised recommenda-
tions for optimal hydration issued by the American College
of Sports Medicine and the International Marathon Medical
Directors Association.13,14 The former guidelines emphasize
preventing dehydration in higher performance athletes by
advice to drink ahead of thirst, and the latter stress drinking
to thirst only to minimize the risk of overhydration, especially
in slower marathon runners and walkers.15
Elevated levels of whole blood lactate were observed in
95% of collapsed runners. Of these, 32 (40.5%) had values
more than the range commonly regarded as a critical value
(31.5 mg/dL [3.5 mmol/L]).16 Of the subjects, 51% had a low
ionized calcium level and 23% an abnormal ionized magne-
sium level (8 of 9 had hypomagnesemia). The low ionized
calcium values are most likely a consequence of the elevations
in lactate because lactic acidosis is known to decrease the ion-
950 Am J Clin Pathol 2008;129:948-951
950 DOI: 10.1309/4L0M60MGAQBCHMV7
ized calcium level.17 Metabolic acidosis also is a known cause
of hypomagnesemia, as was found in 8 cases. Low ionized
calcium and magnesium levels may contribute to impaired
skeletal muscle and, potentially, myocardial function18 in the
collapsed runners. The mean glucose value of 117 mg/dL (6.5
mmol/L) in collapsed runners reported herein is similar to the
mean of 110 mg/dL (6.1 mmol/L) reported in asymptomatic
finishers.6 No collapsed runners were hypoglycemic.
The mechanism underlying exertional triggering of acute
cardiac events remains unknown. Relative risks 2.38- and
16.9-fold greater in apparently healthy women and men,
respectively, have been reported for acute cardiac events,
including sudden death, during strenuous exercise compared
with baseline at rest.1,19 Marathon running is associated with
an abrupt but transient increase in relative risk for ACS in
persons who are otherwise at relatively low risk. Despite
improved emergency cardiac care along race courses in recent
years, the rate of sudden death predominantly in middle aged
men has remained at 1 per 50,000 marathoners during the last
2 decades.4,5 The Marine Corps and Los Angeles marathons
experienced multiple cases of cardiac arrest in 2007, perhaps
reflecting increasing participation of older runners. Boston
Marathon entrants between ages 40 and 59 years increased
from 4,000 in 1997 to 10,000 in 2007.20
Although we show comparable (if not lower) levels of
diagnostic and prognostically meaningful cardiac biomarkers
among endurance athletes compared with levels in athletes
who finished the marathon, we also found prevalent abnor-
malities in various electrolytes, which in conjunction with
lactic acidosis and a decrease in ionized calcium and magne-
sium levels may contribute to the enhanced risk for ventricular
fibrillation during acute exertional events. Furthermore, the
significance of these metabolic abnormalities as related to
exercise-induced collapse is not yet clear.
From the 1Department of Medicine, McLean Hospital, Belmont,
MA; Departments of 2Cardiology and 3Pathology, Massachusetts
General Hospital and 4Harvard Medical School, Boston; and
5Nova Biomedical, Waltham, MA.
Address reprint requests to Dr Lewandrowski: Clinical
Chemistry, Gray 5, Massachusetts General Hospital, Fruit St,
Boston, MA 02114.
References
1. Albert CM, Mittleman MA, Chae CU, et al. Triggering of
sudden death from cardiac causes by vigorous exertion. N Engl
J Med. 2000;343:1355-1361.
2. Thompson PD, Franklin BA, Balady GJ, et al; and the American
Heart Association Council on Nutrition, Physical Activity,
and Metabolism; Council on Clinical Cardiology; American
College of Sports Medicine. Exercise and acute cardiovascular
events placing the risks in perspective: a scientific statement
from the American Heart Association Council on Nutrition,
Physical Activity, and Metabolism and the Council on Clinical
Cardiology. Circulation. 2007;115:2358-2368.
American Society for Clinical Pathology

3. American College of Sports Medicine, American Heart
Association. Exercise and acute cardiovascular events: placing
the risk in perspective. Med Sci Sports Exerc. 2007;39:889-897.
4. Maron BJ, Poliac LC, Roberts WO. Risk for sudden death
associated with marathon running. J Am Coll Cardiol.
1996;28:428-431.
5. Tunstall Pedoe DS. Marathon cardiac deaths: the London
experience. Sports Med. 2007;37:448-450.
6. Kratz A, Lewandrowski K, Siegel A, et al. Effect of running on
hematologic and biochemical laboratory parameters including
cardiac markers. Am J Clin Pathol. 2002;118:856-863.
7. Siegel AJ, Sholar M, Yang J, et al. Elevated serum cardiac
markers in asymptomatic marathon runners after competition:
is the myocardium stunned? Cardiology. 1997;88:487-491.
8. Siegel A, Lewandrowski E, Chun K, et al. Changes in cardiac
markers including B-natriuretic peptide in runners after the
Boston Marathon. Am J Cardiol. 2001;88:920-923.
9. Neilan T, Januzzi J, Lee-Lewandrowski E, et al. Myocardial
injury and ventricular dysfunction related to training levels
among nonelite participants in the Boston Marathon.
Circulation. 2006;114:2325-2333.
10. Middleton N, Shave RE, George K, et al. Impact of repeated
prolonged exercise bouts on cardiac function and biomarkers.
Med Sci Sports Exerc. 2007;39:83-92.
11. Fortescue E, Shin A, Greenes D, et al. Cardiac troponin
increases among runners in the Boston Marathon. Ann Emerg
Med. 2007;49:137-143.
12. Kratz A, Siegel A, Verbalis J, et al. Sodium status of collapsed
marathon runners. Arch Pathol Lab Med. 2005;129:227-230.
American Society for Clinical Pathology
Clinical Chemistry / Original Article
13. Sawka M, Burke L, Eichner R, et al. Exercise and fluid
replacement. Med Sci Sports Exerc. 2007;39:377-389.
14. Maharam LG, Hew T, Siegel AJ, et al. International Marathon
Medical Directors: Revised Fluid Recommendations for
Runners and Walkers. Approved by IMMDA: 6 May 2006,
Barcelona, Spain. Available at: http://www.aims-association.
org/guidelines_fluid_replacement.html. Accessed December 6.
2006.
15. Hew-Butler T, Verbalis J, Noakes T. Updated fluid
recommendation: position statement from the International
Marathon Medical Directors Association (IMMDA). Clin J
Sport Med. 2006;16:283-292.
16. Roberts W, McMillin G, Burtis C, et al. Reference information
for the clinical laboratory. In: Burtis C, Ashwood E, Bruns D,
eds. Tietz Textbook of Clinical Chemistry. 4th ed. Philadelphia,
PA: Elsevier Saunders; 2006:chap 56.
17. Robergs R, Ghiasvand F, Parker D. Biochemistry of exercise-
induced metabolic acidosis. Am J Physiol Regul Intgr Comp
Physiol. 2004;287:R502-R516.
18. Cooper D, Walley K, Dodek P, et al. Plasma ionized calcium
and blood lactate concentrations are inversely associated with
human lactic acidosis. Intensive Care Med. 1992;18:286-289.
19. Whang W, Manson JE, Hu FB, et al. Physical exertion,
exercise, and sudden cardiac death in women. JAMA.
2006;295:1399-1403.
20. History of the Boston Marathon. Available at http://www.
Bostonathleticassociation.org.
Am J Clin Pathol 2008;129:948-951 951
951 DOI: 10.1309/4L0M60MGAQBCHMV7 951