Correspondence
The Hepatopulmonary Syndrome
N Engl J Med 2008; 359:866-867August 21, 2008
- Article
To the Editor:
In their review of the hepatopulmonary syndrome, Rodríguez-Roisin and Krowka (May 29 issue)1 provide a pathophysiological explanation of the symptoms of this syndrome, but they downplay the importance of the history and physical examination in making the diagnosis, stating that there are no signs, symptoms, or hallmarks of the hepatopulmonary syndrome on physical examination. We disagree. Platypnea (the opposite of orthopnea) not only is a common symptom, but also is almost diagnostic in itself in a patient with cirrhosis. Many patients with the hepatopulmonary syndrome also have cyanosis and clubbing, an unusual combination in cirrhosis.
We believe that the combination of known cirrhosis, platypnea, and orthodeoxia that is correctable with oxygen supplementation is absolutely diagnostic. The clinical examination and simple laboratory tests have a critical role in making this diagnosis. Thus, we question the need for expensive scans, since intracardiac shunting can be ruled out if the hypoxemia is responsive to oxygen.
Recognition of platypnea is a practical issue for care. Health care workers almost instinctively try to make patients with dyspnea sit up, but they should learn to let these persons lie down.
1 ReferencesCameron N. Ghent, M.D.
Mark A. Levstik, M.D.
Paul J. Marotta, M.D.
London Health Sciences Centre, London, ON N6A 3R9, Canada
ghent1@on.aibn. com 1
Rodriguez-Roisin R ,Krowka MJ . Hepatopulmonary syndrome -- a liver-induced lung vascular disorder. N Engl J Med 2008;358:2378-2387
Full Text | Web of Science | Medline
To the Editor:
Rodríguez-Roisin and Krowka mention that an alveolar–arterial oxygen gradient of 15 mm Hg or more is one of the diagnostic criteria for defective oxygenation in the hepatopulmonary syndrome. In some earlier studies, the threshold of 20 mm Hg was used.1,2 Since the gradient increases with age, it may be more accurate to compare patients' gradients with those that are corrected for their age (the corrected alveolar–arterial oxygen gradient=10+0.43×[age–20]).3 In a healthy 40-year-old person, the corrected alveolar–arterial oxygen gradient is 18.6 mm Hg, which is already above the lower threshold. In our analysis involving 96 patients on the waiting list for liver transplantation, 40 of the patients (42%) had an alveolar–arterial oxygen gradient above 20 mm Hg.4 If the lower threshold were used, the number of patients would increase to 58 (60%). The hepatopulmonary syndrome was diagnosed in 23 patients (24%), but with a lower threshold, it would have been diagnosed in 33 patients (34%). Therefore, the use of a lower threshold for the alveolar–arterial oxygen gradient, without taking age into consideration, may lead to an overestimation of the incidence of the hepatopulmonary syndrome.
4 ReferencesCezary Szmigielski, M.D., Ph.D.
Rafal Krenke, M.D., Ph.D.
Grzegorz Styczynski, M.D., Ph.D.
Medical University of Warsaw, 02-097 Warsaw, Poland
grzegorz.styczynski@amwaw. edu. pl 1
Swanson KL ,Wiesner RH ,Krowka MJ . Natural history of hepatopulmonary syndrome: impact of liver transplantation. Hepatology 2005;41:1122-1129
CrossRef | Web of Science | Medline2
Krowka MJ . Hepatopulmonary syndromes. Gut 2000;46:1-4
CrossRef | Web of Science | Medline3
Fallon MB ,Abrams GA . Pulmonary dysfunction in chronic liver disease. Hepatology 2000;32:859-865
CrossRef | Web of Science | Medline4
Przybylowski T ,Krenke R ,Fangrat A , et al. Gas exchange abnormalities in patients listed for liver transplantation. J Physiol Pharmacol 2006;57:Suppl 4:313-323
Web of Science | Medline
To the Editor:
Rodríguez-Roisin and Krowka state that contrast-enhanced transthoracic echocardiography cannot distinguish discrete arteriovenous malformations from intracardiac shunt. This technique can, in fact, discern three levels of shunting: atrial septal defect, ventricular septal defect with Eisenmenger's syndrome, and pulmonary arteriovenous malformations.1 In the presence of an atrial septal defect, contrast material is usually recorded in the left atrium within one or two cardiac cycles after its appearance in the right atrium. Late arrival of contrast material in the left atrium after peripheral injection of contrast material (four to eight cardiac cycles after the appearance of contrast in the right atrium) is diagnostic of pulmonary arteriovenous malformations, especially when the contrast material is seen entering from the pulmonary veins. The delayed appearance in the left atrium reflects the time required for passage through the pulmonary circulation. This finding highlights the importance of imaging of the pulmonary veins in identifying the entrance of the bubbles through the veins and not through the interatrial septum.
1 ReferencesJuan F. Viles-Gonzalez, M.D.
Brigham and Women's Hospital, Boston, MA 02115
jviles-gonzalez@partners.org 1
Barzilai B ,Waggoner AD ,Spessert C ,Picus D ,Goodenberger D . Two-dimensional contrast echocardiography in the detection and follow-up of congenital pulmonary arteriovenous malformations. Am J Cardiol 1991;68:1507-1510
CrossRef | Web of Science | Medline
Author/Editor Response
Ghent et al. mention that platypnea in the hepatopulmonary syndrome is a common symptom that is almost diagnostic in itself. We disagree. In a study of the mechanisms of pulmonary gas-exchange abnormalities in 80 candidates for liver transplantation,1 2 of 14 patients with the hepatopulmonary syndrome had platypnea alone, as compared with 11 patients who had nonpositional dyspnea. Also, orthodeoxia is not necessarily correctable with supplemental oxygen, as can be inferred from our article. In 20 patients with the hepatopulmonary syndrome,2 just 2 of 5 patients with orthodeoxia had platypnea, suggesting that these two conditions do not always go together. Platypnea, with or without orthodeoxia, is a rare symptom. In addition to the hepatopulmonary syndrome, platypnea may occur with other causes of major intracardiac shunting such as that due to atrial septal defect, recurrent pulmonary embolism, or the changes after pneumonectomy.3 However, we agree that clinical judgment is necessary in order to unravel the severity of hypoxemia in patients who have the hepatopulmonary syndrome with other pulmonary conditions.
Szmigielski et al. suggest a correction for age in the formula for the calculation of the alveolar–arterial oxygen gradient in order to avoid overdiagnosis of the hepatopulmonary syndrome. We agree, and this is why we proposed the use of an alveolar–arterial oxygen gradient of 20 mm Hg or greater for patients older than 64 years of age (Table 1 of our article). This recommendation is a compromise between the simplest approach, which is to use the threshold of 15 mm Hg, and any of the formulas proposed, and it is based on the study by Schenk et al.4 However, the measurement of the alveolar–arterial oxygen gradient always needs to be considered within the individual clinical context, in which the use of contrast-enhanced transthoracic echocardiography plays a central diagnostic role.
Viles-Gonzalez correctly points out that contrast-enhanced transthoracic echocardiography can distinguish intracardiac from intrapulmonary shunt by timing the clear appearance of microbubbles in the left atrium during the cardiac cycle. These microbubbles appear in less than three cycles in intracardiac shunt and in more than three cycles in intrapulmonary shunt. However, the timing and degree of opacification in the left atrium may not always be clear-cut, or there may be both an intracardiac and intrapulmonary component to the left atrial opacification. We agree with Viles-Gonzalez that in those uncommon situations, the more definitive transesophageal contrast imaging of the pulmonary veins may be required rather than the transthoracic approach.
4 ReferencesRoberto Rodríguez-Roisin, M.D.
University of Barcelona, E-08036 Barcelona, Spain
rororo@clinic.ub. es Michael J. Krowka, M.D.
Mayo Clinic, Rochester, MN 559051
Martinez GP ,Barbera JA ,Visa J , et al. Hepatopulmonary syndrome in candidates for liver transplantation. J Hepatol 2001;34:651-657
CrossRef | Web of Science | Medline2
Gomez FP ,Martinez-Palli G ,Barbera JA ,Roca J ,Navasa M ,Rodriguez-Roisin R . Gas exchange mechanism of orthodeoxia in hepatopulmonary syndrome. Hepatology 2004;40:660-666
CrossRef | Web of Science | Medline3
Faller M ,Kessler R ,Chaouat A ,Ehrhart M ,Petit H ,Weitzenblum E . Platypnea-orthodeoxia syndrome related to an aortic aneurysm combined with an aneurysm of the atrial septum. Chest 2000;118:553-557
CrossRef | Web of Science | Medline4
Schenk P ,Fuhrmann V ,Madl C , et al. Hepatopulmonary syndrome: prevalence and predictive value of various cut offs for arterial oxygenation and their clinical consequences. Gut 2002;51:853-859
CrossRef | Web of Science | Medline
