Respiratory Physiology & Neurobiology 190 (2014) 137–141
Contents lists available at ScienceDirect
Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol
Interstitial lung edema triggered by marathon running Gerald S. Zavorsky a,b,∗ , Eric N.C. Milne c , Federico Lavorini d , Joseph P. Rienzi e , Kaleen M. Lavin f , Allison M. Straub g , Massimo Pistolesi d a
Department of Health and Sport Sciences, University of Louisville, Louisville, KY 40292, United States Department of Physiology and Biophysics, University of Louisville, Louisville, KY 40292, United States Department of Radiological Sciences, University of California – Irvine, Irvine, CA 92697, United States d Section of Respiratory Medicine, Department of Experimental and Clinical Medicine, University of Florence, Florence 50134, Italy e Department of Radiology, Regional Hospital of Scranton, Scranton, PA 18510, United States f Human Performance Laboratory, Ball State University, Muncie, IN 47306, United States g Human Physiology Laboratory, Marywood University, Scranton, PA 18509, United States b
c
a r t i c l e
i n f o
Article history: Accepted 14 October 2013 Keywords: Endurance Exercise Lung fluid Pulmonary Water
a b s t r a c t The purpose of this study was to determine whether marathon running causes lung edema, and if so, to determine its effects on runners. Posterior/anterior (PA) radiographs were taken one day before the marathon and at 19, 55, and 98 min post-marathon in 26 runners. The pre and post exercise radiographs of each runner were collated, and then read simultaneously. Two physicians interpreted the images independently in a blinded fashion. The PA radiographs were viewed together at each time-point and findings suggestive for interstitial lung edema were rated as ‘mild,’ ‘moderate,’ or ‘severe’ based on four different radiological criteria. Forty-six percent of the runners presented radiographic findings suggestive of mild to severe interstitial lung edema. Radiographic findings persisted until 98-min post-marathon, with at least moderate degree increases found more frequently in women (55%) than men (6%) (p < 0.01). In conclusion, about half of the runners developed interstitial lung edema of varying degrees post-exercise with the incidence being higher in women compared to men. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Since 1980, there has been almost a four-fold increase in the total number of marathon competitors in the United States, representing close to 0.2% of the U.S. population over 18 years of age in 2010. Its increasing popularity has been accompanied by a large body of research on the physiology of marathon running. It has been mentioned anecdotally that the normal levels of lung water (40–50 mL of water per L of lung at total lung capacity) (Milne and Pistolesi, 1993) may rise during marathon running. Different types of exercise such as an “all-out” 15-min cycling interval training session (Zavorsky et al., 2006a), a triathlon (Caillaud et al., 1995), an ultramarathon (McKechnie et al., 1979), and sustained heavy cycling exercise for 45 min (McKenzie et al., 2005) caused approximately 65% of subjects to show signs of increased lung water.
∗ Corresponding author at: Department of Health and Sport Sciences, Crawford Gym, Room LL02B, University of Louisville, Louisville, KY 40292, United States. Tel.: +1 502 852 7193; fax: +1 502 852 4534. E-mail addresses: [email protected], [email protected] (G.S. Zavorsky). 1569-9048/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resp.2013.10.007
Despite this evidence, argument continues both for and against the development of lung edema with exercise (Hopkins, 2010a,b; Sheel and McKenzie, 2010a,b). Authors writing about this topic invariably refer to increases in lung water as “edema” which leads to some confusion, since “pulmonary edema” is usually considered condition secondary to an abnormal increase of liquid in the alveoli causing clinical symptoms, whereas smaller increases in lung water can be confined to the interstitial space of the lung. Since in these conditions there is no alteration in the alveolar–capillary interface (Weibel, 1973), the increase does not usually cause any subjective or objective clinical finding. Excess lung water can, therefore, be better detected and defined in its severity by chest radiography rather than by clinical evaluation or by indicator dilution techniques (Pistolesi and Giuntini, 1978). The level of accuracy of chest radiography in detecting and quantifying lung water accumulation has been determined in both animals and humans and has been accepted as the best technique available for this purpose (Pistolesi and Giuntini, 1978; Ware and Matthay, 2005). Although increases in lung water do not cause overt clinical symptoms, we hypothesized that they may have measurable effects on runners’ race performance and post-race fitness. The purpose of this study was to investigate
138
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
the development of radiographic findings suggestive of interstitial lung edema in marathon runners over a range of finishing times. 2. Methods 2.1. Study participants Participants were selected based on projected marathon finishing times, such that about one third of the participants would finish in each of three time ranges: 3:00 h or less, 3:00–4:00 h, and 4:00–5:00 h. The inclusion criteria were as follows: any male or female ≤55 years of age, with the ability to complete the marathon under 5:00 h, and with no symptoms or known presence of heart disease. The exclusion criteria were signs or symptoms of cardiopulmonary disease or obstructive or restrictive lung diseases airways, determined by spirometry testing. The study was approved by the Institutional Review Board of Marywood University. Subjects who agreed to participate signed a written informed consent. 2.2. Imaging technique Each subject had posterior/anterior (PA) radiographs taken prerace, as well as 20, 45 and 90 min after crossing the finish line. All radiographic imaging was performed using a portable X-ray machine (Model SR-115, Source Ray Inc., Ronkonkoma, NY) and Digital Radiology ViZion DR imaging Software/flat panel detectors (Viztek Inc., Garner, NC). The X-ray tube-to-patient distance was six feet. Radiographic exposure was individualized using 90–125 kVp and 2.4 mAs (6 ms) to 9.8 mAs (24 ms) exposure time, depending on the subject’s anthropometric characteristics, and using the lower kV levels to obtain optimum lung contrast. The image was obtained at total lung capacity. Each set provided an effective radiation dose ranging from 0.02 to 0.12 mSV. Thus, the maximum total exposure per subject was 0.48 mSv for the full study. 2.3. X-ray reading The films were analyzed by two readers: a chest radiologist (ENCM) and a chest physician (MP), both of whom have previously published a textbook on chest radiograph reading (Milne and Pistolesi, 1993). Each X-ray set was randomly coded before being supplied to the readers. The readers were blinded as to the time each radiograph was taken. Each reader was given the same set of instructions for interpreting the films and evaluated the films separately from the other readers. The radiographs from each runner were analyzed in a randomized order. Because of the randomization, it was not possible for the readers to know whether they were looking at pre or post exercise films, making it (deliberately) impossible for them to make comparative pre and post film analyses. Films were analyzed for evidence of the following findings (Anholm et al., 1999; Gallagher et al., 1988; Giuntini et al., 1987; Milne, 1985; Milne and Pistolesi, 1993; Miniati et al., 1988; Pistolesi and Giuntini, 1978; Zavorsky et al., 2006a) considered suggestive of interstitial lung edema: (1) loss of sharp definition of pulmonary vascular markings; (2) hilar blurring; (3) peribronchial and perivascular cuffing; (4) obscuration of the smallest peripheral vessels. The first three radiographic findings were graded on a three-point scale: 0 when absent, 1 for minimally present, and 2 if there was a definite radiographic presence. For the fourth finding, this item was scored as 0 for normal visualization, 1 for partly obscured, and 2 for completely obscured. Therefore, the total score for each radiograph could range from 0 (absence of any radiographic findings suggestive of interstitial lung edema) to 8 (maximum total score attainable).
For each subject, the scores of the two readers were then averaged for data analysis. Since the radiographic findings of interstitial lung edema in its early phase of accumulation could be subtle and nonspecific, to avoid false positive readings and increase the specificity of the readings, we arbitrarily graded interstitial lung edema as “none” (total score of 2.0 or lower), “mild” (total score from 2.1 to 3), “moderate” (total score from 3.1 to 4) and “severe” (total score higher than 4). 2.4. Statistical analyses Radiographic characteristics before and after exercise for all runners who finished the race in under 5 h were compared using the Friedman analysis of variance test. This test compared whether the median “edema” scores (ordinal data) amongst the four measurement points for all finishers significantly increased. A Wilcoxon-signed rank test determined where the differences were post hoc. As there were four combinations of pairs (pre vs. 20 min post, pre vs. 45 min post, pre vs. 90 min post, 45 min post vs. 90 min post), p < 0.0125 was used to signify statistical significance. A Kruskal–Wallis ANOVA also compared the differences between edema scores obtained in males compared to females. If significant, a Wilcoxon Mann–Whitney U test then determined where the differences were post hoc. The level of inter-observer agreement on the quantification of radiological findings suggestive of interstitial lung edema was obtained from the weighted kappa statistical test (Jakobsson and Westergren, 2005; Kundel and Polansky, 2003), and the average weighted kappa coefficient was reported (Kundel and Polansky, 2003). Difference between quantification scores for two readers at every time-point was also assessed using a Kruskal–Wallis ANOVA, and when significant, a Wilcoxon Mann–Whitney U test then determined where the differences were post hoc, with significance declared at p < 0.0125. Forward binary logistic regression was conducted to determine which independent variables (marathon finishing time, age, gender) were predictors of developing moderate to severe interstitial pulmonary edema (yes or no) from marathon running. The data were analyzed by a statistical software package (SPSS Version 19.2, IBM SPSS Statistics Inc., Chicago, IL). Statistical significance was declared when p < 0.05 unless otherwise noted. 3. Results The 2011 Steamtown Marathon began at 475 m above sea level and finished at 229 m above sea level. The ambient temperature increased from about 8 ◦ C (96% humidity) at the 8:00 a.m. start to 21 ◦ C (51% humidity) at the finish line by 1:00 p.m. Twenty-six subjects with normal resting lung function values (as reported elsewhere in a study using the same subjects (Lavin et al., 2012)) completed the marathon in less than 5 h with times ranging from 2 h and 22 min to 4 h and 48 min. The average finishing time was 3 h and 38 min (40 min). Of the finishers, 17 were male, with a mean (SD) age of 40 (8) years, height of 178 (6) cm, weight of 76 (10) kg, and body surface area of 1.94 (0.15) m2 . Nine subjects were female and had a mean age of 33 (9) years, height of 164 (8) cm, weight of 57 (9) kg, and body surface area of 1.60 (0.16) m2 . Due to difficulty in organizing flow of runners through the chest X-ray triage, the films were not taken at the exact predetermined times. Instead, the average times for the three post-exercise X-rays were 19 (8) min, 55 (13) min, and 98 (16) min, post-finish, respectively. Fig. 1 shows evidence of moderate to severe findings suggestive of interstitial lung edema in a subject at 58 min post-exercise. Post-marathon, 12 out of the 26 (46%) runners (11 at the first post race time point, and 1 at the second post race time point)
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
139
Fig. 1. Chest radiographs obtained in a representative male subject at pre-race (upper panels) and 58 min after completion of the marathon (lower panels). Lefts panels show full size chest films, right panels show the four criteria suggestive of interstitial lung edema in close up views corresponding to the black outlined area: (1) Large white arrow: The most definitive sign is apparent thickening of the bronchial walls, due to water accumulating in the peribronchovascular sheaths. Compare upper and lower close up views. (2) Black arrow: Loss of sharp definition of vessel margins. Compare upper and lower close up views. (3) Small white arrow: Peribronchial “cuffing”, the bronchus is seen end-on making it very easy to detect. Compare upper and lower close up views. (4) Loss of definition of the smaller peripheral vessels. Compare upper and lower close up views.
developed mild to severe interstitial lung edema at one or more post-exercise time points. Six out of nine (66%) women and five out of 17 (29%) men, (42% overall), had X-ray scores compatible with interstitial lung edema ranging from mild to severe at the first post-race time point (Fig. 2). The overall X-ray scores after marathon completion were significantly higher than baseline at
Fig. 2. Individual lung edema X-ray scores obtained in 17 males (filled circles) and 9 women (empty circles) at baseline (pre race) and 19 (SD 8), 55 (13) and 98 (16) min after completion of the marathon. Horizontal continuous lines indicate mean values. X-ray scores were significantly different from pre-race scores at all three post-race time points (p = 0.001, 0.003, and 0.012, respectively).
19 min (p = 0.001), 55 min (p = 0.003), and 98 min (p = 0.012) postfinish (Fig. 2). Five out of the nine women (55%) and one of seventeen males (6%) were classified as having moderate to severe radiographic findings suggestive of interstitial lung edema at one or more of the post-exercise time-points (Fisher’s exact test between genders, p = 0.009). A difference in X-ray score existed between males and females, but only at 19 min post-exercise (p = 0.015). The mean age between genders tended to be different (p = 0.0531, unpaired t test); mean age did not correlate with the presence of X-ray findings of interstitial lung edema. Males lost significantly more weight (1.3 kg) during running of the marathon compared to females (0.3 kg). The absolute agreement between readers in X-ray scores was 50% and the weighted kappa statistic was 0.60 (95% C.I. 0.47–0.74), indicating moderate agreement among observers. Binary logistic regression revealed that gender was the best predictor of presence of moderate to severe interstitial lung edema (−2 log likelihood = 27.3, Chi-square = 0.01, Nagelkerke r2 = 0.29). A model based on gender alone demonstrated that women appeared to have a higher likelihood of interstitial lung edema than males. Because height and weight were significantly different between genders (p < 0.05), it was hypothesized that differences in incidence of edema might be attributable to differences in body surface area. Gender was strongly associated with body surface area (r = 0.74,
140
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
p = 0.000), and body surface area was associated with the X-ray score post exercise (r = −0.21 to −0.49 at all post-exercise time points). No association was found between gender and marathon finishing time, or between marathon time and X-ray scores. 4. Discussion In this study, we sought to determine whether marathon running triggered interstitial lung edema. We found that 46% of the runners presented X-ray findings suggestive of interstitial lung edema ranging from ‘mild’ to ‘severe’. X-ray findings occurred immediately following the completion of marathon, and, albeit reduced, were still present about one hour and half after completion of the race. Of note, increases of moderate to severe degrees in EVLW were found in a greater proportion of women (55%), compared to men (6%). 4.1. Potential causes of interstitial lung edema One of the potential causes of post-marathon interstitial edema would initially appear to be high pulmonary pressure. However, increased pulmonary blood flow in a normal capillary bed is not likely to cause a rise in pulmonary artery pressure because of the low resistance of the capillary bed and its capacity for recruitment. Alternatively, pulmonary artery pressure may increase secondarily because of a rise in left atrial pressure. It would appear that this is the case in exercise edema when most of the elevation in pulmonary arterial pressure is due to an elevation of mean pulmonary wedge pressure (Harvey et al., 1971). Interestingly, the fitter the individual, the larger the cardiac output during strenuous exercise, and the higher the wedge pressure (Harvey et al., 1971). However, in a small group of subjects, Stickland et al. (2006) found that, compared to less fit subjects, highly-trained subjects had lower pulmonary artery wedge pressure during about 3 min incremental cycle exercise up to 90% of the maximal oxygen consumption, thus suggesting that increased pulmonary driving pressure during short lasting exercise could result from a lower pulmonary artery wedge pressures. Differences in the subjects’ characteristics and in the type and degree of exercise may account for the conflicting results. In a comprehensive literature review of a 47 studies comprising 1187 subjects, Kovacs et al. (2009) reported that wedge pressures during exercise can be as high as 25–30 mmHg, which may be sufficient to cause pulmonary edema. Furthermore, in normal young and old men, exercise has been shown to increase wedge pressures by more than 20 mmHg (Reeves et al., 1996). Additionally, elevated levels of serum cardiac injury markers were observed after marathon in trained runners (Siegel et al., 1997), thus suggesting that post-marathon interstitial edema may arise from cardiac decompensation. A further cause of interstitial edema during exercise could theoretically be ultra-structural mechanical stress caused by increased hydrostatic pressures in the pulmonary capillaries causing leakage of plasma, protein, and red blood cells into the interstitial space (O’Callaghan et al., 1987). However, as shown by West and MathieuCostello (1999) the hydrostatic pressure threshold for such failure typically exceeds that for edema formation and Hopkins et al. (1998) found that sustained heavy exercise in humans did not lead to loss of barrier integrity or red cell accumulation in the airspaces. Furthermore, stress failure would produce “injury edema,” whose radiographic findings are characterized by a patchy, non-gravitational peripheral distribution of edema, typical of alveolar space involvement. These radiographic findings take many days to clear and can be easily differentiated from those occurring in the course of pulmonary interstitial edema caused by increased pulmonary venous pressure (Milne, 1985). All radiographic findings
detected in the runners in this study are in keeping with a hydrostatic origin of interstitial edema. This is also in keeping with the total absence in the runners with post-race X-ray scores compatible with interstitial edema of the clinical findings related to alveolar flooding which are observed in patients with pulmonary edema resulting from injured microvessels. Excess lung water in our runners accumulated to a greater extent in the central perihilar regions of the lungs. This is in keeping with the proposed mechanism that there is a gradient of hydrostatic pressure within the interstitial space of the lung that drives excess liquid filtered at the capillary level toward the loose connective interstitial spaces surrounding the large perihilar vessels and bronchi (Bhattacharya et al., 1984). These spaces act as sumps to preserve the gas exchange function at the alveolar level in the early phase of excess lung water accumulation. The persistence of X-ray findings suggestive of interstitial lung edema one hour and a half after completion of the marathon in only 3 out of the 11 runners who displayed edema at the first post-race time reinforces the hypothesis of an hydrostatic mechanism explaining the occurrence of lung edema after marathon and is in contrast with edema ensuing from injured lung vessels. Another potential cause of pulmonary edema during exercise is alteration in fluid–electrolyte balance in the blood, such that blood becomes hypotonic and fluid diffuses across an osmotic gradient into the brain to compensate for hyponatremia (Ayus et al., 2000). This condition can be brought about during endurance exercise by over-consumption of fluids relative to losses through sweat leading to “water intoxication” (Noakes et al., 2005). Hyponatremic encephalopathy can lead to increased intracranial pressure, which can result in pulmonary edema by two possible mechanisms: (1) centrally mediated increase in pulmonary vascular permeability to proteins, leading to increased alveolar and interstitial fluid (McClellan et al., 1989) and (2) increased sympathetic neuronal activity with catecholamine release, resulting in pulmonary vasoconstriction with increased capillary hydrostatic pressure and capillary wall injury (Smith and Matthay, 1997). A limitation of our study to confirm this hypothesis was that fluid consumption during the race was not measured. However, both mechanisms of pulmonary edema formation secondary to water intoxication and hyponatremic encephalopathy could result in injury lung edema whose radiographic appearance, as already noted, is quite different from that observed in our cases. Furthermore, pulmonary edema caused by exercise is mostly mild interstitial edema (Zavorsky et al., 2006a), and it should not affect pulmonary gas exchange (Zavorsky et al., 2006b). On the contrary, poor pulmonary gas exchange invariably happens with the flooding of the alveolar spaces in the course of injury lung edema. Not one subject in our series of films showed this. 4.2. Limitations It may be argued that the use of chest radiographs to detect and quantify extravascular lung water is a limitation. However, a National Institutes of Health workshop (Staub, 1986) on detection of lung water reported that “the chest X-ray film remains the reference standard against which other lung water methods are compared. Its advantages include moderate accuracy, fair sensitivity, good reproducibility, non-invasiveness, practicality, availability, reliability, portability, ease of use in the emergency care setting, and relatively low cost. It also provides excellent information about edema distribution”. To our knowledge since the publication of this workshop report there have not been relevant methodological changes in the assessment of lung edema for clinical purposes, as it has been confirmed in a more recent review (Ware and Matthay, 2005). The correlation between edema graded by X-ray and quantified by the indicator dilution technique is good (r = 0.83, p < 0.01) and demonstrates that an increase of only 10% in extravascular lung water is detectable
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
with a chest radiograph (Pistolesi and Giuntini, 1978). A second limitation was that the absolute agreement between the two readers for the directional changes in the edema score from pre to post exercise was only 50%. Nevertheless, the calculated weighted kappa statistic between the two readers of 0.60 is classified as “moderate agreement” (Kundel and Polansky, 2003), which demonstrates adequate consistency of these readers in reporting directional changes in scores. 5. Summary In conclusion, 12 out of 26 runners (46%) manifested X-ray signs suggestive of interstitial lung edema ranging from very mild to moderately severe. The occurrence and severity of edema was more likely in females, not related to finishing time, and still observed 1.5 h post-finish, a trend which would have to be confirmed in a future study using more subjects of both sexes. Conflict of interest The authors of the present study have no conflicts of interest. Funding No funding was received for this work. Acknowledgments The authors would like to thank the Steamtown Marathon Race Committee, including Assistant Race Director Jim Cummings, for help in recruiting participants for this study. Additionally, we acknowledge the Steamtown Marathon Medical Director, Tim Rowland, MD, for providing available space at the medical triage area at the finish line. No grant funding was received for this study. This study was presented at the European Respiratory Society Conference, in Vienna, Austria, in September, 2012. The citation for the abstract is: Zavorsky, G.S. Milne, E. N. Pistolesi, M. Lavorini, F., Rienzi, J., Lavin, K.M., Straub, A.M., 2012. Pulmonary edema is frequently triggered by marathon running. [Abstract]. Eur. Respir. J. 40 (Suppl. 56) 281s. References Anholm, J.D., Milne, E.N., Stark, P., Bourne, J.C., Friedman, P., 1999. Radiographic evidence of interstitial pulmonary edema after exercise at altitude. Journal of Applied Physiology 86, 503–509. Ayus, J.C., Varon, J., Arieff, A.I., 2000. Hyponatremia, cerebral edema, and noncardiogenic pulmonary edema in marathon runners. Annals of Internal Medicine 132, 711–714. Bhattacharya, J., Gropper, M.A., Staub, N.C., 1984. Interstitial fluid pressure gradient measured by micropuncture in excised dog lung. Journal of Applied Physiology 56, 271–277. Caillaud, C., Serre-Cousine, O., Anselme, F., Capdevilla, X., Prefaut, C., 1995. Computerized tomography and pulmonary diffusing capacity in highly trained athletes after performing a triathlon. Journal of Applied Physiology 79, 1226–1232. Gallagher, C.G., Huda, W., Rigby, M., Greenberg, D., Younes, M., 1988. Lack of radiographic evidence of interstitial pulmonary edema after maximal exercise in normal subjects. American Review of Respiratory Disease 137, 474–476. Giuntini, C., Pistolesi, M., Miniati, M., Fazio, F., 1987. Extravascular lung water. European Journal of Nuclear Medicine 13 (Suppl), S63–S69. Harvey, R.M., Enson, Y., Ferrer, M.I., 1971. A reconsideration of the origins of pulmonary hypertension. Chest 59, 82–94.
141
Hopkins, S.R., 2010a. Last word on point–counterpoint: pulmonary edema does occur in human athletes performing heavy sea level exercise. Journal of Applied Physiology 109, 1281. Hopkins, S.R., 2010b. Point: pulmonary edema does occur in human athletes performing heavy sea-level exercise. Journal of Applied Physiology 109, 1270–1272. Hopkins, S.R., Schoene, R.B., Henderson, W.R., Spragg, R.G., West, J.B., 1998. Sustained submaximal exercise does not alter the integrity of the lung blood–gas barrier in elite athletes. Journal of Applied Physiology 84, 1185–1189. Jakobsson, U., Westergren, A., 2005. Statistical methods for assessing agreement for ordinal data. Scandinavian Journal of Caring Sciences 19, 427–431. Kovacs, G., Berghold, A., Scheidl, S., Olschewski, H., 2009. Pulmonary arterial pressure during rest and exercise in healthy subjects: a systematic review. European Respiratory Journal 34, 888–894. Kundel, H.L., Polansky, M., 2003. Measurement of observer agreement. Radiology 228, 303–308. Lavin, K.M., Straub, A.M., Uhranowsky, K.A., Smoliga, J.M., Zavorsky, G.S., 2012. Alveolar-membrane diffusing capacity limits performance in Boston marathon qualifiers. PLoS ONE 7, e44513. McClellan, M.D., Dauber, I.M., Weil, J.V., 1989. Elevated intracranial pressure increases pulmonary vascular permeability to protein. Journal of Applied Physiology 67, 1185–1191. McKechnie, J.K., Leary, W.P., Noakes, T.D., Kallmeyer, J.C., MacSearraigh, E.T., Olivier, L.R., 1979. Acute pulmonary oedema in two athletes during a 90-km running race. South African Medical Journal 56, 261–265. McKenzie, D.C., O’Hare, T.J., Mayo, J., 2005. The effect of sustained heavy exercise on the development of pulmonary edema in trained male cyclists. Respiratory Physiology and Neurobiology 145, 209–218. Milne, E.N., 1985. The conventional chest radiograph – does it have a future? Applied Radiology 14, 13–14. Milne, E.N.C., Pistolesi, M., 1993. Detection and Quantification of Pulmonary Edema, In Reading the Chest Radiograph. A Physiologic Approach. Mosby – Year book, Inc., St. Louis, pp. 51–79. Miniati, M., Pistolesi, M., Paoletti, P., Giuntini, C., Lebowitz, M.D., Taylor, A.E., Milne, E.N., 1988. Objective radiographic criteria to differentiate cardiac, renal, and injury lung edema. Investigative Radiology 23, 433–440. Noakes, T.D., Goodwin, N., Rayner, B.L., Branken, T., Taylor, R.K., 2005. Water intoxication: a possible complication during endurance exercise, 1985. Wilderness and Environmental Medicine 16, 221–227. O’Callaghan, M.W., Pascoe, J.R., Tyler, W.S., Mason, D.K., 1987. Exercise-induced pulmonary haemorrhage in the horse: results of a detailed clinical, post mortem and imaging study. VIII. Conclusions and implications. Equine Veterinary Journal 19, 428–434. Pistolesi, M., Giuntini, C., 1978. Assessment of extravascular lung water. Radiology Clinics of North America 16, 551–574. Reeves, J.T., et al., 1996. Pulmonary hemodynamics and fluid exchange in lungs during exercise. In: Handbook of Physiology Section 12: Exercise: Regulation and Integration of Multiple Systems. Oxford University Press, New York, pp. 585–613. Sheel, A.W., McKenzie, D.C., 2010a. Counterpoint: pulmonary edema does not occur in human athletes performing heavy sea-level exercise. Journal of Applied Physiology 109, 1272–1273. Sheel, A.W., McKenzie, D.C., 2010b. Last word on point–counterpoint: pulmonary edema does not occur in human athletes performing heavy sea-level exercise. Journal of Applied Physiology 109, 1282. Siegel, A.J., Sholar, M., Yang, J., Dhanak, E., Lewandrowski, K.B., 1997. Elevated serum cardiac markers in asymptomatic marathon runners after competition: is the myocardium stunned? Cardiology 88, 487–491. Smith, W.S., Matthay, M.A., 1997. Evidence for a hydrostatic mechanism in human neurogenic pulmonary edema. Chest 111, 1326–1333. Staub, N.C., 1986. Clinical use of lung water measurements. Report of a workshop. Chest 90, 588–594. Stickland, M.K., Welsh, R.C., Petersen, S.R., Tyberg, J.V., Anderson, W.D., Jones, R.L., Taylor, D.A., Bouffard, M., Haykowsky, M.J., 2006. Does fitness level modulate the cardiovascular hemodynamic response to exercise? Journal of Applied Physiology 100, 1895–1901. Ware, L.B., Matthay, M.A., 2005. Clinical practice. Acute pulmonary edema. New England Journal of Medicine 353, 2788–2796. Weibel, E.R., 1973. Morphological basis of alveolar–capillary gas exchange. Physiological Reviews 53, 419–495. West, J.B., Mathieu-Costello, O., 1999. Structure, strength, failure, and remodeling of the pulmonary blood–gas barrier. Annual Review of Physiology 61, 543–572. Zavorsky, G.S., Saul, L., Decker, A., Ruiz, P., 2006a. Radiographic evidence of pulmonary edema during high-intensity interval training in women. Respiratory Physiology Neurobiology 153, 181–190. Zavorsky, G.S., Saul, L., Murias, J.M., Ruiz, P., 2006b. Pulmonary gas exchange does not worsen during repeat exercise in women. Respiratory Physiology Neurobiology 153, 226–236.
Contents lists available at ScienceDirect
Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol
Interstitial lung edema triggered by marathon running Gerald S. Zavorsky a,b,∗ , Eric N.C. Milne c , Federico Lavorini d , Joseph P. Rienzi e , Kaleen M. Lavin f , Allison M. Straub g , Massimo Pistolesi d a
Department of Health and Sport Sciences, University of Louisville, Louisville, KY 40292, United States Department of Physiology and Biophysics, University of Louisville, Louisville, KY 40292, United States Department of Radiological Sciences, University of California – Irvine, Irvine, CA 92697, United States d Section of Respiratory Medicine, Department of Experimental and Clinical Medicine, University of Florence, Florence 50134, Italy e Department of Radiology, Regional Hospital of Scranton, Scranton, PA 18510, United States f Human Performance Laboratory, Ball State University, Muncie, IN 47306, United States g Human Physiology Laboratory, Marywood University, Scranton, PA 18509, United States b
c
a r t i c l e
i n f o
Article history: Accepted 14 October 2013 Keywords: Endurance Exercise Lung fluid Pulmonary Water
a b s t r a c t The purpose of this study was to determine whether marathon running causes lung edema, and if so, to determine its effects on runners. Posterior/anterior (PA) radiographs were taken one day before the marathon and at 19, 55, and 98 min post-marathon in 26 runners. The pre and post exercise radiographs of each runner were collated, and then read simultaneously. Two physicians interpreted the images independently in a blinded fashion. The PA radiographs were viewed together at each time-point and findings suggestive for interstitial lung edema were rated as ‘mild,’ ‘moderate,’ or ‘severe’ based on four different radiological criteria. Forty-six percent of the runners presented radiographic findings suggestive of mild to severe interstitial lung edema. Radiographic findings persisted until 98-min post-marathon, with at least moderate degree increases found more frequently in women (55%) than men (6%) (p < 0.01). In conclusion, about half of the runners developed interstitial lung edema of varying degrees post-exercise with the incidence being higher in women compared to men. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Since 1980, there has been almost a four-fold increase in the total number of marathon competitors in the United States, representing close to 0.2% of the U.S. population over 18 years of age in 2010. Its increasing popularity has been accompanied by a large body of research on the physiology of marathon running. It has been mentioned anecdotally that the normal levels of lung water (40–50 mL of water per L of lung at total lung capacity) (Milne and Pistolesi, 1993) may rise during marathon running. Different types of exercise such as an “all-out” 15-min cycling interval training session (Zavorsky et al., 2006a), a triathlon (Caillaud et al., 1995), an ultramarathon (McKechnie et al., 1979), and sustained heavy cycling exercise for 45 min (McKenzie et al., 2005) caused approximately 65% of subjects to show signs of increased lung water.
∗ Corresponding author at: Department of Health and Sport Sciences, Crawford Gym, Room LL02B, University of Louisville, Louisville, KY 40292, United States. Tel.: +1 502 852 7193; fax: +1 502 852 4534. E-mail addresses: [email protected], [email protected] (G.S. Zavorsky). 1569-9048/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resp.2013.10.007
Despite this evidence, argument continues both for and against the development of lung edema with exercise (Hopkins, 2010a,b; Sheel and McKenzie, 2010a,b). Authors writing about this topic invariably refer to increases in lung water as “edema” which leads to some confusion, since “pulmonary edema” is usually considered condition secondary to an abnormal increase of liquid in the alveoli causing clinical symptoms, whereas smaller increases in lung water can be confined to the interstitial space of the lung. Since in these conditions there is no alteration in the alveolar–capillary interface (Weibel, 1973), the increase does not usually cause any subjective or objective clinical finding. Excess lung water can, therefore, be better detected and defined in its severity by chest radiography rather than by clinical evaluation or by indicator dilution techniques (Pistolesi and Giuntini, 1978). The level of accuracy of chest radiography in detecting and quantifying lung water accumulation has been determined in both animals and humans and has been accepted as the best technique available for this purpose (Pistolesi and Giuntini, 1978; Ware and Matthay, 2005). Although increases in lung water do not cause overt clinical symptoms, we hypothesized that they may have measurable effects on runners’ race performance and post-race fitness. The purpose of this study was to investigate
138
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
the development of radiographic findings suggestive of interstitial lung edema in marathon runners over a range of finishing times. 2. Methods 2.1. Study participants Participants were selected based on projected marathon finishing times, such that about one third of the participants would finish in each of three time ranges: 3:00 h or less, 3:00–4:00 h, and 4:00–5:00 h. The inclusion criteria were as follows: any male or female ≤55 years of age, with the ability to complete the marathon under 5:00 h, and with no symptoms or known presence of heart disease. The exclusion criteria were signs or symptoms of cardiopulmonary disease or obstructive or restrictive lung diseases airways, determined by spirometry testing. The study was approved by the Institutional Review Board of Marywood University. Subjects who agreed to participate signed a written informed consent. 2.2. Imaging technique Each subject had posterior/anterior (PA) radiographs taken prerace, as well as 20, 45 and 90 min after crossing the finish line. All radiographic imaging was performed using a portable X-ray machine (Model SR-115, Source Ray Inc., Ronkonkoma, NY) and Digital Radiology ViZion DR imaging Software/flat panel detectors (Viztek Inc., Garner, NC). The X-ray tube-to-patient distance was six feet. Radiographic exposure was individualized using 90–125 kVp and 2.4 mAs (6 ms) to 9.8 mAs (24 ms) exposure time, depending on the subject’s anthropometric characteristics, and using the lower kV levels to obtain optimum lung contrast. The image was obtained at total lung capacity. Each set provided an effective radiation dose ranging from 0.02 to 0.12 mSV. Thus, the maximum total exposure per subject was 0.48 mSv for the full study. 2.3. X-ray reading The films were analyzed by two readers: a chest radiologist (ENCM) and a chest physician (MP), both of whom have previously published a textbook on chest radiograph reading (Milne and Pistolesi, 1993). Each X-ray set was randomly coded before being supplied to the readers. The readers were blinded as to the time each radiograph was taken. Each reader was given the same set of instructions for interpreting the films and evaluated the films separately from the other readers. The radiographs from each runner were analyzed in a randomized order. Because of the randomization, it was not possible for the readers to know whether they were looking at pre or post exercise films, making it (deliberately) impossible for them to make comparative pre and post film analyses. Films were analyzed for evidence of the following findings (Anholm et al., 1999; Gallagher et al., 1988; Giuntini et al., 1987; Milne, 1985; Milne and Pistolesi, 1993; Miniati et al., 1988; Pistolesi and Giuntini, 1978; Zavorsky et al., 2006a) considered suggestive of interstitial lung edema: (1) loss of sharp definition of pulmonary vascular markings; (2) hilar blurring; (3) peribronchial and perivascular cuffing; (4) obscuration of the smallest peripheral vessels. The first three radiographic findings were graded on a three-point scale: 0 when absent, 1 for minimally present, and 2 if there was a definite radiographic presence. For the fourth finding, this item was scored as 0 for normal visualization, 1 for partly obscured, and 2 for completely obscured. Therefore, the total score for each radiograph could range from 0 (absence of any radiographic findings suggestive of interstitial lung edema) to 8 (maximum total score attainable).
For each subject, the scores of the two readers were then averaged for data analysis. Since the radiographic findings of interstitial lung edema in its early phase of accumulation could be subtle and nonspecific, to avoid false positive readings and increase the specificity of the readings, we arbitrarily graded interstitial lung edema as “none” (total score of 2.0 or lower), “mild” (total score from 2.1 to 3), “moderate” (total score from 3.1 to 4) and “severe” (total score higher than 4). 2.4. Statistical analyses Radiographic characteristics before and after exercise for all runners who finished the race in under 5 h were compared using the Friedman analysis of variance test. This test compared whether the median “edema” scores (ordinal data) amongst the four measurement points for all finishers significantly increased. A Wilcoxon-signed rank test determined where the differences were post hoc. As there were four combinations of pairs (pre vs. 20 min post, pre vs. 45 min post, pre vs. 90 min post, 45 min post vs. 90 min post), p < 0.0125 was used to signify statistical significance. A Kruskal–Wallis ANOVA also compared the differences between edema scores obtained in males compared to females. If significant, a Wilcoxon Mann–Whitney U test then determined where the differences were post hoc. The level of inter-observer agreement on the quantification of radiological findings suggestive of interstitial lung edema was obtained from the weighted kappa statistical test (Jakobsson and Westergren, 2005; Kundel and Polansky, 2003), and the average weighted kappa coefficient was reported (Kundel and Polansky, 2003). Difference between quantification scores for two readers at every time-point was also assessed using a Kruskal–Wallis ANOVA, and when significant, a Wilcoxon Mann–Whitney U test then determined where the differences were post hoc, with significance declared at p < 0.0125. Forward binary logistic regression was conducted to determine which independent variables (marathon finishing time, age, gender) were predictors of developing moderate to severe interstitial pulmonary edema (yes or no) from marathon running. The data were analyzed by a statistical software package (SPSS Version 19.2, IBM SPSS Statistics Inc., Chicago, IL). Statistical significance was declared when p < 0.05 unless otherwise noted. 3. Results The 2011 Steamtown Marathon began at 475 m above sea level and finished at 229 m above sea level. The ambient temperature increased from about 8 ◦ C (96% humidity) at the 8:00 a.m. start to 21 ◦ C (51% humidity) at the finish line by 1:00 p.m. Twenty-six subjects with normal resting lung function values (as reported elsewhere in a study using the same subjects (Lavin et al., 2012)) completed the marathon in less than 5 h with times ranging from 2 h and 22 min to 4 h and 48 min. The average finishing time was 3 h and 38 min (40 min). Of the finishers, 17 were male, with a mean (SD) age of 40 (8) years, height of 178 (6) cm, weight of 76 (10) kg, and body surface area of 1.94 (0.15) m2 . Nine subjects were female and had a mean age of 33 (9) years, height of 164 (8) cm, weight of 57 (9) kg, and body surface area of 1.60 (0.16) m2 . Due to difficulty in organizing flow of runners through the chest X-ray triage, the films were not taken at the exact predetermined times. Instead, the average times for the three post-exercise X-rays were 19 (8) min, 55 (13) min, and 98 (16) min, post-finish, respectively. Fig. 1 shows evidence of moderate to severe findings suggestive of interstitial lung edema in a subject at 58 min post-exercise. Post-marathon, 12 out of the 26 (46%) runners (11 at the first post race time point, and 1 at the second post race time point)
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
139
Fig. 1. Chest radiographs obtained in a representative male subject at pre-race (upper panels) and 58 min after completion of the marathon (lower panels). Lefts panels show full size chest films, right panels show the four criteria suggestive of interstitial lung edema in close up views corresponding to the black outlined area: (1) Large white arrow: The most definitive sign is apparent thickening of the bronchial walls, due to water accumulating in the peribronchovascular sheaths. Compare upper and lower close up views. (2) Black arrow: Loss of sharp definition of vessel margins. Compare upper and lower close up views. (3) Small white arrow: Peribronchial “cuffing”, the bronchus is seen end-on making it very easy to detect. Compare upper and lower close up views. (4) Loss of definition of the smaller peripheral vessels. Compare upper and lower close up views.
developed mild to severe interstitial lung edema at one or more post-exercise time points. Six out of nine (66%) women and five out of 17 (29%) men, (42% overall), had X-ray scores compatible with interstitial lung edema ranging from mild to severe at the first post-race time point (Fig. 2). The overall X-ray scores after marathon completion were significantly higher than baseline at
Fig. 2. Individual lung edema X-ray scores obtained in 17 males (filled circles) and 9 women (empty circles) at baseline (pre race) and 19 (SD 8), 55 (13) and 98 (16) min after completion of the marathon. Horizontal continuous lines indicate mean values. X-ray scores were significantly different from pre-race scores at all three post-race time points (p = 0.001, 0.003, and 0.012, respectively).
19 min (p = 0.001), 55 min (p = 0.003), and 98 min (p = 0.012) postfinish (Fig. 2). Five out of the nine women (55%) and one of seventeen males (6%) were classified as having moderate to severe radiographic findings suggestive of interstitial lung edema at one or more of the post-exercise time-points (Fisher’s exact test between genders, p = 0.009). A difference in X-ray score existed between males and females, but only at 19 min post-exercise (p = 0.015). The mean age between genders tended to be different (p = 0.0531, unpaired t test); mean age did not correlate with the presence of X-ray findings of interstitial lung edema. Males lost significantly more weight (1.3 kg) during running of the marathon compared to females (0.3 kg). The absolute agreement between readers in X-ray scores was 50% and the weighted kappa statistic was 0.60 (95% C.I. 0.47–0.74), indicating moderate agreement among observers. Binary logistic regression revealed that gender was the best predictor of presence of moderate to severe interstitial lung edema (−2 log likelihood = 27.3, Chi-square = 0.01, Nagelkerke r2 = 0.29). A model based on gender alone demonstrated that women appeared to have a higher likelihood of interstitial lung edema than males. Because height and weight were significantly different between genders (p < 0.05), it was hypothesized that differences in incidence of edema might be attributable to differences in body surface area. Gender was strongly associated with body surface area (r = 0.74,
140
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
p = 0.000), and body surface area was associated with the X-ray score post exercise (r = −0.21 to −0.49 at all post-exercise time points). No association was found between gender and marathon finishing time, or between marathon time and X-ray scores. 4. Discussion In this study, we sought to determine whether marathon running triggered interstitial lung edema. We found that 46% of the runners presented X-ray findings suggestive of interstitial lung edema ranging from ‘mild’ to ‘severe’. X-ray findings occurred immediately following the completion of marathon, and, albeit reduced, were still present about one hour and half after completion of the race. Of note, increases of moderate to severe degrees in EVLW were found in a greater proportion of women (55%), compared to men (6%). 4.1. Potential causes of interstitial lung edema One of the potential causes of post-marathon interstitial edema would initially appear to be high pulmonary pressure. However, increased pulmonary blood flow in a normal capillary bed is not likely to cause a rise in pulmonary artery pressure because of the low resistance of the capillary bed and its capacity for recruitment. Alternatively, pulmonary artery pressure may increase secondarily because of a rise in left atrial pressure. It would appear that this is the case in exercise edema when most of the elevation in pulmonary arterial pressure is due to an elevation of mean pulmonary wedge pressure (Harvey et al., 1971). Interestingly, the fitter the individual, the larger the cardiac output during strenuous exercise, and the higher the wedge pressure (Harvey et al., 1971). However, in a small group of subjects, Stickland et al. (2006) found that, compared to less fit subjects, highly-trained subjects had lower pulmonary artery wedge pressure during about 3 min incremental cycle exercise up to 90% of the maximal oxygen consumption, thus suggesting that increased pulmonary driving pressure during short lasting exercise could result from a lower pulmonary artery wedge pressures. Differences in the subjects’ characteristics and in the type and degree of exercise may account for the conflicting results. In a comprehensive literature review of a 47 studies comprising 1187 subjects, Kovacs et al. (2009) reported that wedge pressures during exercise can be as high as 25–30 mmHg, which may be sufficient to cause pulmonary edema. Furthermore, in normal young and old men, exercise has been shown to increase wedge pressures by more than 20 mmHg (Reeves et al., 1996). Additionally, elevated levels of serum cardiac injury markers were observed after marathon in trained runners (Siegel et al., 1997), thus suggesting that post-marathon interstitial edema may arise from cardiac decompensation. A further cause of interstitial edema during exercise could theoretically be ultra-structural mechanical stress caused by increased hydrostatic pressures in the pulmonary capillaries causing leakage of plasma, protein, and red blood cells into the interstitial space (O’Callaghan et al., 1987). However, as shown by West and MathieuCostello (1999) the hydrostatic pressure threshold for such failure typically exceeds that for edema formation and Hopkins et al. (1998) found that sustained heavy exercise in humans did not lead to loss of barrier integrity or red cell accumulation in the airspaces. Furthermore, stress failure would produce “injury edema,” whose radiographic findings are characterized by a patchy, non-gravitational peripheral distribution of edema, typical of alveolar space involvement. These radiographic findings take many days to clear and can be easily differentiated from those occurring in the course of pulmonary interstitial edema caused by increased pulmonary venous pressure (Milne, 1985). All radiographic findings
detected in the runners in this study are in keeping with a hydrostatic origin of interstitial edema. This is also in keeping with the total absence in the runners with post-race X-ray scores compatible with interstitial edema of the clinical findings related to alveolar flooding which are observed in patients with pulmonary edema resulting from injured microvessels. Excess lung water in our runners accumulated to a greater extent in the central perihilar regions of the lungs. This is in keeping with the proposed mechanism that there is a gradient of hydrostatic pressure within the interstitial space of the lung that drives excess liquid filtered at the capillary level toward the loose connective interstitial spaces surrounding the large perihilar vessels and bronchi (Bhattacharya et al., 1984). These spaces act as sumps to preserve the gas exchange function at the alveolar level in the early phase of excess lung water accumulation. The persistence of X-ray findings suggestive of interstitial lung edema one hour and a half after completion of the marathon in only 3 out of the 11 runners who displayed edema at the first post-race time reinforces the hypothesis of an hydrostatic mechanism explaining the occurrence of lung edema after marathon and is in contrast with edema ensuing from injured lung vessels. Another potential cause of pulmonary edema during exercise is alteration in fluid–electrolyte balance in the blood, such that blood becomes hypotonic and fluid diffuses across an osmotic gradient into the brain to compensate for hyponatremia (Ayus et al., 2000). This condition can be brought about during endurance exercise by over-consumption of fluids relative to losses through sweat leading to “water intoxication” (Noakes et al., 2005). Hyponatremic encephalopathy can lead to increased intracranial pressure, which can result in pulmonary edema by two possible mechanisms: (1) centrally mediated increase in pulmonary vascular permeability to proteins, leading to increased alveolar and interstitial fluid (McClellan et al., 1989) and (2) increased sympathetic neuronal activity with catecholamine release, resulting in pulmonary vasoconstriction with increased capillary hydrostatic pressure and capillary wall injury (Smith and Matthay, 1997). A limitation of our study to confirm this hypothesis was that fluid consumption during the race was not measured. However, both mechanisms of pulmonary edema formation secondary to water intoxication and hyponatremic encephalopathy could result in injury lung edema whose radiographic appearance, as already noted, is quite different from that observed in our cases. Furthermore, pulmonary edema caused by exercise is mostly mild interstitial edema (Zavorsky et al., 2006a), and it should not affect pulmonary gas exchange (Zavorsky et al., 2006b). On the contrary, poor pulmonary gas exchange invariably happens with the flooding of the alveolar spaces in the course of injury lung edema. Not one subject in our series of films showed this. 4.2. Limitations It may be argued that the use of chest radiographs to detect and quantify extravascular lung water is a limitation. However, a National Institutes of Health workshop (Staub, 1986) on detection of lung water reported that “the chest X-ray film remains the reference standard against which other lung water methods are compared. Its advantages include moderate accuracy, fair sensitivity, good reproducibility, non-invasiveness, practicality, availability, reliability, portability, ease of use in the emergency care setting, and relatively low cost. It also provides excellent information about edema distribution”. To our knowledge since the publication of this workshop report there have not been relevant methodological changes in the assessment of lung edema for clinical purposes, as it has been confirmed in a more recent review (Ware and Matthay, 2005). The correlation between edema graded by X-ray and quantified by the indicator dilution technique is good (r = 0.83, p < 0.01) and demonstrates that an increase of only 10% in extravascular lung water is detectable
G.S. Zavorsky et al. / Respiratory Physiology & Neurobiology 190 (2014) 137–141
with a chest radiograph (Pistolesi and Giuntini, 1978). A second limitation was that the absolute agreement between the two readers for the directional changes in the edema score from pre to post exercise was only 50%. Nevertheless, the calculated weighted kappa statistic between the two readers of 0.60 is classified as “moderate agreement” (Kundel and Polansky, 2003), which demonstrates adequate consistency of these readers in reporting directional changes in scores. 5. Summary In conclusion, 12 out of 26 runners (46%) manifested X-ray signs suggestive of interstitial lung edema ranging from very mild to moderately severe. The occurrence and severity of edema was more likely in females, not related to finishing time, and still observed 1.5 h post-finish, a trend which would have to be confirmed in a future study using more subjects of both sexes. Conflict of interest The authors of the present study have no conflicts of interest. Funding No funding was received for this work. Acknowledgments The authors would like to thank the Steamtown Marathon Race Committee, including Assistant Race Director Jim Cummings, for help in recruiting participants for this study. Additionally, we acknowledge the Steamtown Marathon Medical Director, Tim Rowland, MD, for providing available space at the medical triage area at the finish line. No grant funding was received for this study. This study was presented at the European Respiratory Society Conference, in Vienna, Austria, in September, 2012. The citation for the abstract is: Zavorsky, G.S. Milne, E. N. Pistolesi, M. Lavorini, F., Rienzi, J., Lavin, K.M., Straub, A.M., 2012. Pulmonary edema is frequently triggered by marathon running. [Abstract]. Eur. Respir. J. 40 (Suppl. 56) 281s. References Anholm, J.D., Milne, E.N., Stark, P., Bourne, J.C., Friedman, P., 1999. Radiographic evidence of interstitial pulmonary edema after exercise at altitude. Journal of Applied Physiology 86, 503–509. Ayus, J.C., Varon, J., Arieff, A.I., 2000. Hyponatremia, cerebral edema, and noncardiogenic pulmonary edema in marathon runners. Annals of Internal Medicine 132, 711–714. Bhattacharya, J., Gropper, M.A., Staub, N.C., 1984. Interstitial fluid pressure gradient measured by micropuncture in excised dog lung. Journal of Applied Physiology 56, 271–277. Caillaud, C., Serre-Cousine, O., Anselme, F., Capdevilla, X., Prefaut, C., 1995. Computerized tomography and pulmonary diffusing capacity in highly trained athletes after performing a triathlon. Journal of Applied Physiology 79, 1226–1232. Gallagher, C.G., Huda, W., Rigby, M., Greenberg, D., Younes, M., 1988. Lack of radiographic evidence of interstitial pulmonary edema after maximal exercise in normal subjects. American Review of Respiratory Disease 137, 474–476. Giuntini, C., Pistolesi, M., Miniati, M., Fazio, F., 1987. Extravascular lung water. European Journal of Nuclear Medicine 13 (Suppl), S63–S69. Harvey, R.M., Enson, Y., Ferrer, M.I., 1971. A reconsideration of the origins of pulmonary hypertension. Chest 59, 82–94.
141
Hopkins, S.R., 2010a. Last word on point–counterpoint: pulmonary edema does occur in human athletes performing heavy sea level exercise. Journal of Applied Physiology 109, 1281. Hopkins, S.R., 2010b. Point: pulmonary edema does occur in human athletes performing heavy sea-level exercise. Journal of Applied Physiology 109, 1270–1272. Hopkins, S.R., Schoene, R.B., Henderson, W.R., Spragg, R.G., West, J.B., 1998. Sustained submaximal exercise does not alter the integrity of the lung blood–gas barrier in elite athletes. Journal of Applied Physiology 84, 1185–1189. Jakobsson, U., Westergren, A., 2005. Statistical methods for assessing agreement for ordinal data. Scandinavian Journal of Caring Sciences 19, 427–431. Kovacs, G., Berghold, A., Scheidl, S., Olschewski, H., 2009. Pulmonary arterial pressure during rest and exercise in healthy subjects: a systematic review. European Respiratory Journal 34, 888–894. Kundel, H.L., Polansky, M., 2003. Measurement of observer agreement. Radiology 228, 303–308. Lavin, K.M., Straub, A.M., Uhranowsky, K.A., Smoliga, J.M., Zavorsky, G.S., 2012. Alveolar-membrane diffusing capacity limits performance in Boston marathon qualifiers. PLoS ONE 7, e44513. McClellan, M.D., Dauber, I.M., Weil, J.V., 1989. Elevated intracranial pressure increases pulmonary vascular permeability to protein. Journal of Applied Physiology 67, 1185–1191. McKechnie, J.K., Leary, W.P., Noakes, T.D., Kallmeyer, J.C., MacSearraigh, E.T., Olivier, L.R., 1979. Acute pulmonary oedema in two athletes during a 90-km running race. South African Medical Journal 56, 261–265. McKenzie, D.C., O’Hare, T.J., Mayo, J., 2005. The effect of sustained heavy exercise on the development of pulmonary edema in trained male cyclists. Respiratory Physiology and Neurobiology 145, 209–218. Milne, E.N., 1985. The conventional chest radiograph – does it have a future? Applied Radiology 14, 13–14. Milne, E.N.C., Pistolesi, M., 1993. Detection and Quantification of Pulmonary Edema, In Reading the Chest Radiograph. A Physiologic Approach. Mosby – Year book, Inc., St. Louis, pp. 51–79. Miniati, M., Pistolesi, M., Paoletti, P., Giuntini, C., Lebowitz, M.D., Taylor, A.E., Milne, E.N., 1988. Objective radiographic criteria to differentiate cardiac, renal, and injury lung edema. Investigative Radiology 23, 433–440. Noakes, T.D., Goodwin, N., Rayner, B.L., Branken, T., Taylor, R.K., 2005. Water intoxication: a possible complication during endurance exercise, 1985. Wilderness and Environmental Medicine 16, 221–227. O’Callaghan, M.W., Pascoe, J.R., Tyler, W.S., Mason, D.K., 1987. Exercise-induced pulmonary haemorrhage in the horse: results of a detailed clinical, post mortem and imaging study. VIII. Conclusions and implications. Equine Veterinary Journal 19, 428–434. Pistolesi, M., Giuntini, C., 1978. Assessment of extravascular lung water. Radiology Clinics of North America 16, 551–574. Reeves, J.T., et al., 1996. Pulmonary hemodynamics and fluid exchange in lungs during exercise. In: Handbook of Physiology Section 12: Exercise: Regulation and Integration of Multiple Systems. Oxford University Press, New York, pp. 585–613. Sheel, A.W., McKenzie, D.C., 2010a. Counterpoint: pulmonary edema does not occur in human athletes performing heavy sea-level exercise. Journal of Applied Physiology 109, 1272–1273. Sheel, A.W., McKenzie, D.C., 2010b. Last word on point–counterpoint: pulmonary edema does not occur in human athletes performing heavy sea-level exercise. Journal of Applied Physiology 109, 1282. Siegel, A.J., Sholar, M., Yang, J., Dhanak, E., Lewandrowski, K.B., 1997. Elevated serum cardiac markers in asymptomatic marathon runners after competition: is the myocardium stunned? Cardiology 88, 487–491. Smith, W.S., Matthay, M.A., 1997. Evidence for a hydrostatic mechanism in human neurogenic pulmonary edema. Chest 111, 1326–1333. Staub, N.C., 1986. Clinical use of lung water measurements. Report of a workshop. Chest 90, 588–594. Stickland, M.K., Welsh, R.C., Petersen, S.R., Tyberg, J.V., Anderson, W.D., Jones, R.L., Taylor, D.A., Bouffard, M., Haykowsky, M.J., 2006. Does fitness level modulate the cardiovascular hemodynamic response to exercise? Journal of Applied Physiology 100, 1895–1901. Ware, L.B., Matthay, M.A., 2005. Clinical practice. Acute pulmonary edema. New England Journal of Medicine 353, 2788–2796. Weibel, E.R., 1973. Morphological basis of alveolar–capillary gas exchange. Physiological Reviews 53, 419–495. West, J.B., Mathieu-Costello, O., 1999. Structure, strength, failure, and remodeling of the pulmonary blood–gas barrier. Annual Review of Physiology 61, 543–572. Zavorsky, G.S., Saul, L., Decker, A., Ruiz, P., 2006a. Radiographic evidence of pulmonary edema during high-intensity interval training in women. Respiratory Physiology Neurobiology 153, 181–190. Zavorsky, G.S., Saul, L., Murias, J.M., Ruiz, P., 2006b. Pulmonary gas exchange does not worsen during repeat exercise in women. Respiratory Physiology Neurobiology 153, 226–236.
