Swimming-Induced Pulmonary Edema
2020; Elsevier BV; Volume: 158; Issue: 4 Linguagem: Inglês
10.1016/j.chest.2020.04.028
ISSN1931-3543
AutoresMaria Hårdstedt, Claudia Seiler, Linda Kristiansson, Daniel Lundeqvist, Cecilia Klingberg, Annika Eriksson,
Tópico(s)Cardiac Arrest and Resuscitation
ResumoBackgroundDespite the increasing popularity of open water swimming worldwide, swimming-induced pulmonary edema (SIPE) is a poorly recognized condition lacking established diagnostic criteria.Research QuestionThe aim of this study was to identify diagnostic criteria of SIPE during a large open water swimming event.Study Design and MethodsIn this cross-sectional study, 17,904 individuals swam 1,000, 1,500, or 3,000 m in cold open water during Sweden's largest open water swimming event in 2018 and 2019. Of 166 swimmers seeking medical attention for acute respiratory symptoms, 160 were included in the study. Medical history, symptoms, and clinical findings were collected. On-site lung ultrasound (LUS) was performed to verify pulmonary edema.ResultsPulmonary edema was confirmed by LUS in 102 patients (64%); findings were unilateral in 11 (7%). Peripheral oxygen saturation was identified as a strong independent diagnostic test for pulmonary edema, with ≤ 95% as the suggested cut off based on receiver-operating characteristic curve analysis (area under the curve, 0.893; P < .0001). Crackles on lung auscultation, predominantly over the anterior chest, identified 88% of patients with edema. Peripheral oxygen saturation ≤ 95% or auscultation findings of crackles identified pulmonary edema with a sensitivity of 97% and a specificity of 86%. A specificity of 98% and a positive predictive value of 99% for LUS-verified pulmonary edema were reached if patients presented with both oxygen saturation ≤ 95% and auscultation of crackles.InterpretationWe suggest a clinical algorithm for diagnosis of SIPE for swimmers with acute respiratory symptoms during swimming in cold open water. Novel features of focally distributed edema in the anterior parts of the lungs, sometimes unilateral, add to this unique dataset of an underreported condition. Despite the increasing popularity of open water swimming worldwide, swimming-induced pulmonary edema (SIPE) is a poorly recognized condition lacking established diagnostic criteria. The aim of this study was to identify diagnostic criteria of SIPE during a large open water swimming event. In this cross-sectional study, 17,904 individuals swam 1,000, 1,500, or 3,000 m in cold open water during Sweden's largest open water swimming event in 2018 and 2019. Of 166 swimmers seeking medical attention for acute respiratory symptoms, 160 were included in the study. Medical history, symptoms, and clinical findings were collected. On-site lung ultrasound (LUS) was performed to verify pulmonary edema. Pulmonary edema was confirmed by LUS in 102 patients (64%); findings were unilateral in 11 (7%). Peripheral oxygen saturation was identified as a strong independent diagnostic test for pulmonary edema, with ≤ 95% as the suggested cut off based on receiver-operating characteristic curve analysis (area under the curve, 0.893; P < .0001). Crackles on lung auscultation, predominantly over the anterior chest, identified 88% of patients with edema. Peripheral oxygen saturation ≤ 95% or auscultation findings of crackles identified pulmonary edema with a sensitivity of 97% and a specificity of 86%. A specificity of 98% and a positive predictive value of 99% for LUS-verified pulmonary edema were reached if patients presented with both oxygen saturation ≤ 95% and auscultation of crackles. We suggest a clinical algorithm for diagnosis of SIPE for swimmers with acute respiratory symptoms during swimming in cold open water. Novel features of focally distributed edema in the anterior parts of the lungs, sometimes unilateral, add to this unique dataset of an underreported condition. Take-home PointA clinical algorithm is presented for assessment of patients with acute respiratory symptoms and suspected swimming-induced pulmonary edema during open water swimming. The algorithm identifies pulmonary edema, verified by lung ultrasound, based on peripheral oxygen saturation and lung auscultation.FOR EDITORIAL COMMENT, SEE PAGE 1329Immersion pulmonary edema was first described in divers in 1981 and later also reported among open water swimmers as swimming-induced pulmonary edema (SIPE).1Wilmshurst P. Nuri M. Crowther A. Betts J. Webb-Peploe N. Forearm vacsular response in subjects who develop recurrent pulmonary oedema when scuba diving: a new syndrome.Br Heart J. 1981; 45Google Scholar,2Wilmshurst P.T. Nuri M. Crowther A. Webb-Peploe M.M. Cold-induced pulmonary oedema in scuba divers and swimmers and subsequent development of hypertension.Lancet. 1989; 1: 62-65Abstract PubMed Scopus (148) Google Scholar The pathophysiology of SIPE is not fully understood, but central pooling of blood during immersion in cold water in combination with heavy exercise contribute to increased pulmonary vascular pressure and subsequent pulmonary edema.3Bove A.A. Pulmonary aspects of exercise and sports.Methodist DeBakey Cardiovascular J. 2016; 12: 93-97Crossref PubMed Scopus (7) Google Scholar SIPE predominantly occurs in otherwise healthy individuals.2Wilmshurst P.T. Nuri M. Crowther A. Webb-Peploe M.M. Cold-induced pulmonary oedema in scuba divers and swimmers and subsequent development of hypertension.Lancet. 1989; 1: 62-65Abstract PubMed Scopus (148) Google Scholar,4Shupak A. Weiler-Ravell D. Adir Y. Daskalovic Y.I. Ramon Y. Kerem D. Pulmonary oedema induced by strenuous swimming: a field study.Respir Physiol. 2000; 121: 25-31Crossref PubMed Scopus (66) Google Scholar,5Weiler-Ravell D. Shupak A. Goldenberg I. et al.Pulmonary oedema and haemoptysis induced by strenuous swimming.BMJ. 1995; 311: 361-362Crossref PubMed Scopus (103) Google Scholar The condition usually resolves spontaneously within 48 h, but life-threatening cases in swimmers and fatalities in divers have been reported.4Shupak A. Weiler-Ravell D. Adir Y. Daskalovic Y.I. Ramon Y. Kerem D. Pulmonary oedema induced by strenuous swimming: a field study.Respir Physiol. 2000; 121: 25-31Crossref PubMed Scopus (66) Google Scholar, 5Weiler-Ravell D. Shupak A. Goldenberg I. et al.Pulmonary oedema and haemoptysis induced by strenuous swimming.BMJ. 1995; 311: 361-362Crossref PubMed Scopus (103) Google Scholar, 6Grunig H. Nikolaidis P.T. Moon R.E. Knechtle B. Diagnosis of swimming induced pulmonary edema—a review.Frontiers Physiol. 2017; 8: 652Crossref Scopus (32) Google Scholar, 7Cochard G. Arvieux J. Lacour J.M. Madouas G. Mongredien H. Arvieux C.C. Pulmonary edema in scuba divers: recurrence and fatal outcome.Undersea Hyperb Med. 2005; 32: 39-44PubMed Google Scholar SIPE is characterized by acute onset of cough and dyspnea, sometimes accompanied by excessive sputum and hemoptysis, when swimming in open water.6Grunig H. Nikolaidis P.T. Moon R.E. Knechtle B. Diagnosis of swimming induced pulmonary edema—a review.Frontiers Physiol. 2017; 8: 652Crossref Scopus (32) Google Scholar,8Hohmann E. Glatt V. Tetsworth K. Swimming induced pulmonary oedema in athletes—a systematic review and best evidence synthesis.BMC Sports Sci Med Rehabil. 2018; 10: 18Crossref PubMed Scopus (13) Google Scholar Findings of crackles or wheezes/rhonchi on pulmonary auscultation, together with desaturation, are often described. Diagnostic criteria for SIPE vary in published cases, from a merely descriptive definition based on symptoms, to well-defined clinical parameters and radiologic verification of pulmonary edema.4Shupak A. Weiler-Ravell D. Adir Y. Daskalovic Y.I. Ramon Y. Kerem D. Pulmonary oedema induced by strenuous swimming: a field study.Respir Physiol. 2000; 121: 25-31Crossref PubMed Scopus (66) Google Scholar,6Grunig H. Nikolaidis P.T. Moon R.E. Knechtle B. Diagnosis of swimming induced pulmonary edema—a review.Frontiers Physiol. 2017; 8: 652Crossref Scopus (32) Google Scholar,9Adir Y. Shupak A. Gil A. et al.Swimming-induced pulmonary edema: clinical presentation and serial lung function.Chest. 2004; 126: 394-399Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar,10Ludwig B.B. Mahon R.T. Schwartzman E.L. Cardiopulmonary function after recovery from swimming-induced pulmonary edema.Clin J Sport Med. 2006; 16: 348-351Crossref PubMed Scopus (32) Google Scholar Reports of the condition are mainly case series of young athletic men and triathletes; large cohort studies are lacking.4Shupak A. Weiler-Ravell D. Adir Y. Daskalovic Y.I. Ramon Y. Kerem D. Pulmonary oedema induced by strenuous swimming: a field study.Respir Physiol. 2000; 121: 25-31Crossref PubMed Scopus (66) Google Scholar,5Weiler-Ravell D. Shupak A. Goldenberg I. et al.Pulmonary oedema and haemoptysis induced by strenuous swimming.BMJ. 1995; 311: 361-362Crossref PubMed Scopus (103) Google Scholar,9Adir Y. Shupak A. Gil A. et al.Swimming-induced pulmonary edema: clinical presentation and serial lung function.Chest. 2004; 126: 394-399Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar With the rising popularity of open water swimming among recreational swimmers as well as professional athletes, increased knowledge on SIPE is crucial.8Hohmann E. Glatt V. Tetsworth K. Swimming induced pulmonary oedema in athletes—a systematic review and best evidence synthesis.BMC Sports Sci Med Rehabil. 2018; 10: 18Crossref PubMed Scopus (13) Google Scholar,11Spencer S. Dickinson J. Forbes L. Occurrence, risk factors, prognosis and prevention of swimming-induced pulmonary oedema: a systematic review.Sports Med Open. 2018; 4: 43Crossref PubMed Scopus (13) Google Scholar, 12Smith R. Ormerod J.O.M. Sabharwal N. Kipps C. Swimming-induced pulmonary edema: current perspectives.Open Access J Sports Med. 2018; 9: 131-137Crossref PubMed Google Scholar, 13Kumar M. Thompson P.D. A literature review of immersion pulmonary edema.Phys Sportsmed. 2019; 47: 148-151Crossref PubMed Scopus (10) Google Scholar, 14Braman Eriksson A. Annsberg M. Hardstedt M. Swimming-induced pulmonary edema in Swedish conditions has been insufficiently studied [in Swedish].Lakartidningen. 2017; 114PubMed Google Scholar A clinical algorithm is presented for assessment of patients with acute respiratory symptoms and suspected swimming-induced pulmonary edema during open water swimming. The algorithm identifies pulmonary edema, verified by lung ultrasound, based on peripheral oxygen saturation and lung auscultation. FOR EDITORIAL COMMENT, SEE PAGE 1329 To evaluate diagnostic criteria for SIPE, we studied 160 swimmers with respiratory symptoms in a large cohort of 17,904 individuals participating in Sweden's largest annual open water event. Our aim was to identify easily assessed clinical criteria based on physical examination and presentation of symptoms. Lung ultrasound (LUS) was performed on-site to confirm pulmonary edema. LUS is a valuable tool in the differential diagnostics of acute respiratory failure and is highly sensitive for detection of extrapulmonary lung water.15Lichtenstein D. Lung ultrasound in the critically ill.Curr Opin Crit Care. 2014; 20: 315-322Crossref PubMed Scopus (127) Google Scholar, 16Enghard P. Rademacher S. Nee J. et al.Simplified lung ultrasound protocol shows excellent prediction of extravascular lung water in ventilated intensive care patients.Critical Care. 2015; 19: 36Crossref PubMed Scopus (116) Google Scholar, 17Pivetta E. Goffi A. Lupia E. et al.Lung ultrasound-implemented diagnosis of acute decompensated heart failure in the ED: a SIMEU multicenter study.Chest. 2015; 148: 202-210Abstract Full Text Full Text PDF PubMed Scopus (240) Google Scholar In our context, LUS was also an effective alternative to other radiologic equipment on-site during an open water event.18Winkler M.H. Touw H.R. van de Ven P.M. Twisk J. Tuinman P.R. Diagnostic accuracy of chest radiograph, and when concomitantly studied lung ultrasound, in critically ill patients with respiratory symptoms: a systematic review and meta-analysis.Crit Care Med. 2018; 46: e707-e714Crossref PubMed Scopus (68) Google Scholar Vansbrosimningen is the largest swimming event in Sweden, attracting approximately 10,000 swimmers yearly during a 3-day weekend in July. It takes place in the municipality of Vansbro, 239 m above sea level, in cold freshwater rivers (15°C-20°C). Participants (10 years or older), of both sexes, athlete swimmers as well as unexperienced swimmers swim distances of 1,000, 1,500, or 3,000 m. A vast majority of the swimmers (98% among the included patients) wore wetsuits. The on-site health care consists of a first aid team along the riverside and a mobile medical unit (MMU) at the finish area. Because Vansbro is located 78 km from the nearest hospital, all patients with medical needs were primarily treated at the MMU on site. The total numbers of participants (age ≥ 18 years) at Vansbrosimningen were 11,010 in 2018 and 9,811 in 2019. Of these, 2,917 individuals participated in the event both years. All patients aged ≥ 18 years seeking help at the MMU with acute onset of cough and/or dyspnea commencing during or immediately following swimming were eligible for the study. Medical history, information about the race, symptoms, and clinical findings were collected within approximately 10 to 60 min following swimming. Symptoms were reported as "cough," "dyspnea," "increased sputum," and/or "hemoptysis." Patients were asked if they had aspirated water during the race. Lung auscultation findings were reported as "normal breathing sounds," "crackles," "rhonchi," or "other findings." The location of auscultation findings (right/left chest, anterior/posterior chest) was noted. Peripheral oxygen saturation (Spo2) and heart rate were measured by using a pulse oximeter (Nellcor OxiMax N-65 and Nellcor PM10N; Covidien) and BP using a digital automatic BP monitor (Omron HEM-907; Omron Healthcare Co. Ltd.). Spo2 was noted after the pulse oximeters showed a waveform curve or gave a signal for a synchronized pulse registration. Ethical approval was received from the regional ethical review board in Uppsala, Sweden (Dnr 2017/216 and 2017/216/1). All patients provided written informed consent for participation. All patients with respiratory symptoms at the MMU were examined with LUS by two experienced consultant anesthesiologists. Ultrasound machines aimed for point-of-care examinations with a lung preset adjusted to a depth of 12 cm were used (BK Medical Flex Focus 500 with a curved probe [BK medical type 8823] of 2-6 MHz; BK Medical AB). Patients, sitting in upright position, were scanned for four left chest regions and four right chest regions: upper and basal anterior, and upper and basal lateral.19Volpicelli G. Elbarbary M. Blaivas M. et al.International evidence-based recommendations for point-of-care lung ultrasound.Intensive Care Med. 2012; 38: 577-591Crossref PubMed Scopus (1726) Google Scholar For each lung region, a clip of 2 s was recorded and later independently reviewed by another physician in a blinded manner. The presence of three or more B-line artifacts (B-lines) confirmed a positive region. Bilateral or unilateral presence of two or more positive regions of the four anterior/lateral regions was defined as pulmonary edema in the current cohort, and these patients were referred to as "LUS positive (+)." Patients lacking this pattern of B-lines were further referred to as "LUS negative (–)." In 2019, two posterior regions (upper and basal posterior) were additionally scanned on each side for all patients. Posterior regions were not included when defining pulmonary edema according to LUS. The χ2 test was used for comparisons of categorical outcomes. For continuous data, the unpaired Student t test or the Mann-Whitney U test were used for comparisons between two groups. The Wilcoxon signed-rank test evaluated differences between two groups of related samples. Agreement between tests is presented as percentages and Cohen's kappa coefficient (к). The receiver-operating characteristic curve was used to illustrate the diagnostic value of Spo2. Multiple logistic regression analysis was performed to valuate diagnostic criteria. IBM SPSS Statistics for Windows, version 25.0 (IBM SPSS Statistics, IBM Corporation), and GraphPad Prism version 7.00 (GraphPad Software) were used for statistical analysis and graphic presentation. P values < .05 were considered statistically significant. For comparisons of clinical findings between LUS (+) and LUS (–) patients, a P value < .006 was considered statistically significant based on Bonferroni correction due to multiple testing. Altogether, 166 individuals matched the inclusion criteria. Five patients were excluded due to missing LUS data, and two patients declined participation. We included 160 admissions for 158 individuals; two patients, who sought medical attention at two different occasions, were included twice. In 2018, a reference group of 48 swimmers (aged ≥ 18 years) completing the same swimming race without respiratory symptoms was recruited. Of 158 patients (160 admissions), 102 were positive on LUS examination, and 58 were negative. No pathologic artifacts other than B-lines were identified on LUS. Pulmonary edema according to LUS was predominantly detected over the anterior compared with the lateral regions (P < .0001) (Fig 1A). The median number of positive regions for LUS (+) patients was seven of eight regions examined (interquartile range, 2). The interobserver agreement for LUS evaluation based on all regions (n = 2,004) was 97.5% with a κ value of 0.94 (95% CI, 0.92-0.96). Eleven patients with respiratory symptoms were examined with B-lines predominantly in one lung, as a sign of unilateral edema. Seven of these patients had a distinct unilateral edema with three to four positive regions. In nine of 11 patients with unilateral pulmonary edema, the right lung was affected. In the reference group of 48 swimmers without cough or dyspnea, one person had bilateral pulmonary edema on LUS examination, and one had unilateral pulmonary edema. Scanning of posterior regions in 2019 showed that B-line artifacts were rare: only one posterior sector for one patient was positive among all 85 patients screened. Baseline characteristics for LUS (+) and LUS (–) patients, as well as the reference group, are presented in Table 1. The distance swum was similar in both groups; however, more LUS (+) patients discontinued the race (P < .0001). The group with LUS-verified pulmonary edema was on average older than the group without edema (47 vs 38 years; P < .0001). The most common self-reported co-morbidities were asthma or hypertension in both LUS (+) and LUS (–) patients. Four patients reported heart disease (two reported cardiac murmurs of unknown significance). None had previously known ischemic heart disease or heart failure.Table 1Background Characteristics and Clinical Findings for LUS (+) and LUS (–) Patients With Respiratory Symptoms and for a Reference Group With No Respiratory SymptomsVariablePatients With Cough and/or DyspneaReference Group (n = 48)P ValuebComparison between the reference group and LUS (+) and LUS (–) patients; Student t test or Mann-Whitney U test for continuous variables; χ2 test or Fisher exact test for categorical variablesLUS (+) (n = 102)LUS (–) (n = 58)P ValueaComparison between LUS (+) and LUS (–) patients; Student t test or Mann-Whitney U test for continuous variables; χ2 test or Fisher exact test for categorical variables. The level for significance was set to .006 after Bonferroni correction for multiple comparisons.Baseline characteristics Age, mean ± SD, y47 ± 1038 ± 1242 ± 15 Sex, No. (%)Male8 (8)10 (17)15 (31)Female94 (92)48 (83)33 (69) Medical history, No. (%)Smoker1 (1)2 (3)1 (2)Respiratory disease19 (19)21 (36)3 (6)Asthma18 (18)20 (34)3 (6)Heart disease3 (3)1 (2)3 (6)Hypertension11 (11)4 (7)1 (2) Factors concerning the race, No. (%)Distance1,000 m43 (42)16 (28)8 (17)1,500 m12 (12)9 (15)1 (2)3,000 m47 (46)33 (57)39 (81)Discontinuation of the race45 (44)5 (9)0Clinical findings Symptoms, No. (%)Aspiration during the race45c (45)30c (53).32912 (25).006Cough79 (77)43 (74).636Dyspnea77 (75)33 (57).015Increased sputum32 (31)6 (10).003Hemoptysis3 (3)0NA Clinical findingsCyanosis, No. (%)20e (20)1 (2)NA1d (2)NASpo2, %92 (6)98 (2)< .000198d (2)< .0001Respiratory rate, breaths/min20f (6)20g (8).75820e (7).155Systolic BP, mm Hg129d (25)125 (25)e.292127 (17).979Heart rate, beats/min96c (22)107 (23)< .0001101 (19).596 Lung auscultation, No. (%)Normal10 (10)47 (81)46 (96)Crackles only86 (84)6 (10)2 (4)Crackles + rhonchi4 (4)00Rhonchi only2 (2)4 (7)0Other findings01 (2)Median and interquartile range (IQR) are given if not otherwise stated. LUS = lung ultrasound; LUS (+) = patients with pulmonary edema according to LUS; LUS (–) = patients with no pulmonary edema according to LUS; NA = not applicable due to small sample size; Spo2 = peripheral oxygen saturation.Missing values: cn =1; dn = 2; en = 3; fn = 6; gn = 7.a Comparison between LUS (+) and LUS (–) patients; Student t test or Mann-Whitney U test for continuous variables; χ2 test or Fisher exact test for categorical variables. The level for significance was set to .006 after Bonferroni correction for multiple comparisons.b Comparison between the reference group and LUS (+) and LUS (–) patients; Student t test or Mann-Whitney U test for continuous variables; χ2 test or Fisher exact test for categorical variables Open table in a new tab Median and interquartile range (IQR) are given if not otherwise stated. LUS = lung ultrasound; LUS (+) = patients with pulmonary edema according to LUS; LUS (–) = patients with no pulmonary edema according to LUS; NA = not applicable due to small sample size; Spo2 = peripheral oxygen saturation. Missing values: cn =1; dn = 2; en = 3; fn = 6; gn = 7. A majority of patients in both the LUS (+) and LUS (–) groups reported cough and/or dyspnea as the only symptom (Fig 2, Table 1). Although cough was reported to the same extent in both groups, dyspnea showed a tendency to be more often reported by patients with pulmonary edema (P = .015). Increased sputum and/or hemoptysis was associated with findings of pulmonary edema (P = .003). None of the patients was reported as disoriented at admission. The occurrence of self-reported water aspiration was similar in the LUS (+) and LUS (–) groups (P = .33). None of the included patients had an episode of severe aspiration. Three LUS (+) patients and six LUS (–) patients presented with symptoms of chest pain (3% and 10%, respectively). Pathologic lung auscultation findings were reported for 90% of LUS (+) patients and 19% of LUS (–) patients. For LUS (+) patients, these findings were reported as crackles (84%), rhonchi (2%), or both crackles and rhonchi (4%). Crackles had a higher specificity (90%) and positive predictive value (94%) for pulmonary edema according to LUS compared with including also the findings of rhonchi only (data not shown). Of six patients presenting with rhonchi only, two were LUS (+), and both of these patients had an Spo2 < 95% at admission. Crackles were noted in the majority of LUS (+) patients (84%; n = 86) heard over the anterior lung fields and in 46% (n = 47) exclusively over the anterior lung fields (Fig 1B). The median Spo2 in the LUS (+) group was 92% compared with 98% in the LUS (–) group (P < .0001) (Table 1). Only three individuals in the LUS (–) group had an Spo2 ≤ 95% (95%, 92%, and 85%, respectively); the patient with Spo2 of 85% had previously known asthma. In the reference group, 46 of 48 individuals had normal lung auscultation findings, and the Spo2 was on average 98% after completing the race. Spo2 was a strong diagnostic test for pulmonary edema based on receiver-operating characteristic curve analysis (area under the curve, 0.893; 95% CI, 0.838-0.948; P < .0001) (Fig 3). The optimal cut off level of ≤ 95% in Spo2 for LUS edema was based on sensitivity and specificity tradeoff (Youden index, 176). Using the finding of pulmonary edema according to LUS as outcome, the diagnostic value of the following clinical criteria was assessed: (1) Spo2 ≤ 95%; (2) crackles on lung auscultation; and (3) sputum and/or hemoptysis (Fig 4, Tables 2 and 3). Saturation ≤ 95% and auscultation of crackles were independently associated with findings of pulmonary edema with ORs of 40 and 65, respectively, in a multiple logistic regression model; the model was adjusted for age. Despite difference in frequencies of sputum and/or hemoptysis between LUS (+) and LUS (–) patients (Table 1), occurrence of sputum and/or hemoptysis failed as a predictor of LUS-verified pulmonary edema in the adjusted model. Interestingly, all patients presenting with sputum and/or hemoptysis in addition to auscultation of crackles or Spo2 ≤ 95% had pulmonary edema on LUS.Figure 4Tree diagram presenting clinical findings in patients with acute onset of cough and/or dyspnea during or immediately following Vansbrosimningen in 2018 and 2019. Patients sorted stepwise based on Spo2 ≤ 95% or > 95%, lung auscultation findings of crackles, and reported symptoms of excessive S/H. Number of LUS (+) and LUS (–) patients are presented for each group. S/H = sputum and/or hemoptysis. See Figure 1 and 2 legends for expansion of abbreviations.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table 2Multiple Logistic Regression Analysis of Diagnostic Criteria for the Outcome of Pulmonary Edema Based on LUSVariableParameter EstimateAdjusted OR95% CIP ValueIntercept–7.14Spo2 ≤ 95%3.6839.627.52-208.82< .0001Crackles4.1764.5813.10-318.44< .0001Sputum/hemoptysis0.591.810.27-11.87.539The regression model was adjusted by age. Overall percentage predicted by model, 94.4%. Nagelkerke R2 = 0.84. SIPE = swimming-induced pulmonary edema. See Table 1 legend for expansion of other abbreviations. Open table in a new tab Table 3Evaluation of Diagnostic Criteria for Pulmonary Edema based on LUSLUS (+)(n = 102)LUS (–)(n = 58)Sensitivity (%) (95% CI)Specificity (%)(95% CI)PPV (%)(95% CI)NPV (%)(95% CI)Clinical criteria SaturationSpo2 ≤ 95%84382 (75-90)95 (89-100)97 (93-100)75 (65-85)Spo2 > 95%1855 AuscultationCrackles90688 (82-94)90 (82-97)94 (89-99)81 (72-91)No crackles1252 SymptomsSputum and/or hemoptysis33632 (23-41)90 (82-97)85 (73-96)43 (34-52)Cough and/or dyspnea only6952Combination of clinical criteria Spo2 ≤ 95% or cracklesYes99897 (94-100)86 (77-95)93 (88-98)94 (88-100)No350 Spo2 ≤ 95% and cracklesYes75174 (65-82)98 (95-100)99 (96-100)68 (58-78)No2757Criteria include Spo2 with a cut off of 95%, auscultation findings of crackles, and reported symptoms of sputum and/or hemoptysis. Suggested combination of clinical criteria for SIPE are also evaluated. NPV = negative predictive value; PPV = positive predictive value. See Table 1 and 2 legends for expansion of other abbreviations. Open table in a new tab The regression model was adjusted by age. Overall percentage predicted by model, 94.4%. Nagelkerke R2 = 0.84. SIPE = swimming-induced pulmonary edema. See Table 1 legend for expansion of other abbreviations. Criteria include Spo2 with a cut off of 95%, auscultation findings of crackles, and reported symptoms of sputum and/or hemoptysis. Suggested combination of clinical criteria for SIPE are also evaluated. NPV = negative predictive value; PPV = positive predictive value. See Table 1 and 2 legends for expansion of other abbreviations. A diagnostic algorithm for SIPE diagnosis was suggested (Fig 5). Clinical findings of Spo2 ≤ 95% or auscultation of crackles identified LUS-verified pulmonary edema with a sensitivity of 97% and a specificity of 86% (Table 3). Based on this algorithm, all but three patients with LUS-verified edema were identified (Fig 4). A combination of both these clinical criteria increased the specificity for LUS (+) to 98% with a subsequent decrease in sensitivity to 74%. To the best of our knowledge, this study is the first structured evaluation of clinical diagnosis of SIPE in open water swimmers with acute respiratory symptoms, based on objective verification of pulmonary edema (Fig 5). This is the largest group of swimmers presented thus far with LUS-verified pulmonary edema during an open water swimming event. There are a few reports of pulmonary edema verified by LUS in breath-hold divers, but only one previous case report on SIPE.20Frassi F. Pingitore A. Cialoni D. Picano E. Chest sonography detects lung water accumulation in healthy elite apnea divers.J Am Soc Echocardiogr. 2008; 21: 1150-1155Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar, 21Boussuges A. Coulange M. Bessereau J. et al.Ultrasound lung comets induced by repeated breath-hold diving, a study in underwater fishermen.Scand J Med Sci Sports. 2011; 21: e384-e392Crossref PubMed Scopus (15) Google Scholar, 22Alonso J.V. Chowdhury M. Borakati R. Gankande U. Swimming-induced pulmonary oedema an uncommon condition diagnosed with POCUS ultrasound.Am J Emerg Med. 2017; 35: 1986.e3-1986.e4Abstract Full Text Full Text PDF Scopus (6) Google Scholar A strength of the study is the generalizability based on a large heterogeneous population of swimmers, including all ages, both sexes, and various levels of swimming skills. In addition to acute onset of characteristic symptoms (cough, dyspnea, excessive sputum, and/or hemoptysis) during swimming in open water, we suggest two clinical criteria for SIPE diagnosis: Spo2 ≤ 95% and/or lung auscultation findings of crackles (Fig 5). These two criteria identify pulmonary edema with a high sensitivity based on LUS examination. Based on our data, the occurrence of pulmonary edema in a group with acute respiratory symptoms during swimming in cold open water is high, and the suggested algorithm will obtain a desired high negative predictive value at the first clinical evaluation. In c
Referência(s)