Correspondence
Pulmonary Dead Space and Survival
N Engl J Med 2002; 347:850-852September 12, 2002
- Article
To the Editor:
Nuckton and associates (April 25 issue)1 report an abnormally high dead-space fraction (the ratio of dead-space ventilation to tidal volume, or VD/VT) early in the clinical course of the acute respiratory distress syndrome. Their inference of a heretofore unrecognized functional abnormality is unwarranted, because VD/VT was assessed on the basis of carbon dioxide excretion with the usual Bohr–Enghoff equation. By definition, the patients had refractory hypoxemia, which reflects a severe imbalance between alveolar ventilation (VA) and perfusion (Q), leading to a reduction in the efficiency of intrapulmonary gas transfer for both oxygen and carbon dioxide.2 Thus, assessment of VD/VT on the basis of carbon dioxide excretion is sensitive to any imbalance in VA/Q — notably, in the midrange of values for VA/Q — and even shunt, although to a lesser extent.3,4 This fact depends on the carbon dioxide blood–air partition coefficient's being relatively close to the normal range of VA/Q.4 Specific assessment of high VA/Q areas and “true” dead space is possible, although practical only in small numbers of subjects, with the multiple inert gas elimination technique. We found that in eight patients with the acute respiratory distress syndrome, the mean (±SD) value for VD/VT as measured on the basis of carbon dioxide excretion was considerably higher (0.64±0.03) than the value as measured with the multiple inert gas elimination technique (0.37±0.04).5 In only one of these subjects did we detect areas with high values for VA/Q (range, 10 to 100). That VD/VT was more closely correlated with the outcome than was the ratio of the partial pressure of oxygen (PaO2) to the fraction of inspired oxygen (FiO2) remains a novel and highly relevant finding of the study by Nuckton et al. that may be due to the greater influence on the latter index of factors not intrinsic to the lung — notably, FiO2 and mixed venous oxygen content.
5 ReferencesFrançois Feihl, M.D.
University Hospital, 1011 Lausanne, Switzerland
francois.feihl@chuv. hospvd. ch Christian Melot, M.D.
Serge Brimioulle, M.D.
Erasme University Hospital, 1070 Brussels, Belgium1
Nuckton TJ ,Alonso JA ,Kallet RH , et al. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 2002;346:1281-1286
Full Text | Web of Science | Medline2
West JB . State of the art: ventilation-perfusion relationships. Am Rev Respir Dis 1977;116:919-943
Web of Science | Medline3
Coffey RL ,Albert RK ,Robertson HT . Mechanisms of physiological dead space response to PEEP after acute oleic acid lung injury. J Appl Physiol 1983;55:1550-1557
Web of Science | Medline4
Hlastala MP ,Robertson HT . Inert gas elimination characteristics of the normal and abnormal lung. J Appl Physiol 1978;44:258-266
Web of Science | Medline5
Feihl F ,Eckert P ,Brimioulle S , et al. Permissive hypercapnia impairs pulmonary gas exchange in the adult respiratory distress syndrome. Am J Respir Crit Care Med 2000;162:209-215
Web of Science | Medline
To the Editor:
Nuckton et al. report an association between an elevated physiological dead-space fraction and a high mortality rate among patients with the acute respiratory distress syndrome. Although these results are consistent with those I have reported,1 I am concerned that the method the authors used to measure dead-space fraction is inaccurate. In their report, the effect of compressed volume on the measurement of the dead-space fraction was poorly addressed. In the patients who died, as compared with those who survived, quasistatic respiratory compliance was much lower, indicating that more volume was compressed in the ventilator, and hence more dilution of mixed gases took place, resulting in an overestimation of the dead-space fraction. In some patients, the authors tried to correct this effect by applying a correction factor. Instead, the compressed volume should have been separated by using a Frumin valve.2
Also, the effect of intrapulmonary shunt on the measurement of the dead-space fraction was completely ignored. In the presence of intrapulmonary shunt, Enghoff's modification of the Bohr equation overestimates the dead-space fraction.3 Since the ratio of PaO2 to FiO2 was higher in the patients who died, indicating higher shunt, an overestimation of the dead-space fraction would be more significant. Considering that the mean difference in the dead-space fraction between the two groups was only 0.09, which was the same as 1 SD, correction for the two effects, gas compression and intrapulmonary shunt, may nullify the difference. Nevertheless, in acute respiratory failure, right ventricular function is the chief determinant of the dead-space fraction.1,4
4 ReferencesCharles Her, M.D.
New York Medical College, Valhalla, NY 105951
Her C . Right ventricular stroke-work: an index of distribution of pulmonary perfusion in acute respiratory failure. Chest 1983;84:719-724
CrossRef | Web of Science | Medline2
Forbat AF ,Her C . Correction for gas compression in mechanical ventilators. Anesth Analg 1980;59:488-493
CrossRef | Web of Science | Medline3
Kuwabar S ,Duncalf D . Effect of anatomic shunt on physiologic deadspace-to-tidal volume ratio: a new equation. Anesthesiology 1969;31:575-577
CrossRef | Web of Science | Medline4
Her C ,Lees DE . Accurate assessment of right ventricular function in acute respiratory failure. Crit Care Med 1993;21:1665-1672
CrossRef | Web of Science | Medline
To the Editor:
The study performed by Nuckton et al. was a prospective cohort study. They report an alarming escalation in the risk of death, an odds ratio of 1.45 (95 percent confidence interval, 1.15 to 1.83), for every 0.05 increase in the dead-space fraction.
Odds ratios are increasingly being reported because of widespread use of logistic regression. However, the odds ratio is a difficult concept, and clinicians often interpret it as an approximation of the relative risk. Although it is best to think of the odds ratio as a measure in its own right, the relative risk should be reported whenever it is feasible to do so, especially when the prevalence of the outcome of interest in the unexposed group is high.1,2 In fact, except in case–control studies, the relative risk can be estimated in most studies.3,4
With the use of the method suggested by Zhang and Yu,4 the relative risk and its 95 percent confidence interval can be estimated on the basis of the odds ratio generated from a logistic-regression model: relative risk = odds ratio ÷ [(1 – PO) + (PO × odds ratio)], where PO represents the prevalence of the outcome of interest in the unexposed group.
The prevalence of death in the cohort studied by Nuckton et al. was fairly high (42 percent). With the use of the above equation, the relative risk of death in this study would be 1.22 (95 percent confidence interval, 1.08 to 1.36) for every 0.05 increase in the dead-space fraction. This figure is significantly lower than the odds ratio reported by Nuckton et al. and provides a clearer interpretation of the actual relative risk associated with an elevated dead-space fraction in patients with the acute respiratory distress syndrome than does the odds ratio.
4 ReferencesKwok M. Ho, M.B., B.S.
North Shore Hospital, Auckland 1309, New Zealand
mingk@whl.co. nz 1
Deeks J . When can odds ratios mislead? Odds ratios should be used only in case-control studies and logistic regression analyses. BMJ 1998;317:1155-1156
CrossRef | Web of Science | Medline2
Altman DG ,Deeks JJ ,Sackett DL . Odds ratios should be avoided when events are common. BMJ 1998;317:1318-1318[Erratum, BMJ 1998;317:1505.]
CrossRef | Web of Science | Medline3
Davies HT ,Crombie IK ,Tavakoli M . When can odds ratios mislead? BMJ 1998;316:989-991
CrossRef | Web of Science | Medline4
Zhang J ,Yu KF . What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998;280:1690-1691
CrossRef | Web of Science | Medline
To the Editor:
Nuckton et al. demonstrate the use of measurements of pulmonary dead space in patients with the acute respiratory distress syndrome as a disease-specific index of risk stratification. Although injury to the pulmonary microcirculation with resulting thrombosis has been well described in this syndrome,1 another mechanism for increased dead space is low cardiac output. As cardiac output declines, the fall in lung perfusion yields lung units that are ventilated without perfusion. In the setting of mechanical ventilation, cardiac output is often reduced because of decreased venous return caused by positive intrathoracic pressure. Thus, an elevated dead-space fraction may be a reflection of inadequate volume resuscitation. Increased pulmonary dead space early in the acute respiratory distress syndrome should prompt a search for the underlying cause (e.g., excessive extrinsic or intrinsic positive end-expiratory pressure, relative hypovolemia, microcirculatory thrombosis, or pulmonary embolus), with therapy targeted to the identified cause. This measurement could be used as a simple screening technique early in the course of the acute respiratory distress syndrome to identify patients who may benefit from interventions. Maximization of cardiac output with early aggressive resuscitation can lead to a significant reduction in mortality.2 Late in the disease, however, increases in oxygen delivery are probably unhelpful, because of mitochondrial failure in tissues.3,4 This simple, safe, and inexpensive measurement deserves further study.
4 ReferencesSanjay R. Patel, M.D.
R. Scott Harris, M.D.
Atul Malhotra, M.D.
Harvard Medical School, Boston, MA 02115
spatel@partners.org 1
Tomashefski JF Jr ,Davies P ,Boggis C ,Greene R ,Zapol WM ,Reid LM . The pulmonary vascular lesions of the adult respiratory distress syndrome. Am J Pathol 1983;112:112-126
Web of Science | Medline2
Rivers E ,Nguyen B ,Havstad S , et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368-1377
Full Text | Web of Science | Medline3
Schwartz DR ,Malhotra A ,Fink MP . Cytopathic hypoxia in sepsis: an overview. Sepsis 1999;2:279-289
CrossRef4
Gattinoni L ,Brazzi L ,Pelosi P , et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. N Engl J Med 1995;333:1025-1032
Full Text | Web of Science | Medline
To the Editor:
Nuckton et al. calculated the dead-space fraction by measuring the mean expired carbon dioxide fraction with the use of a bedside metabolic monitor, at a standard tidal volume of 10 ml per kilogram of ideal body weight. Tidal volumes for some patients were altered for 10 minutes before and during the period of measurement. We wonder why the authors used a tidal volume of 10 ml per kilogram. It is troubling, since the use of a tidal volume of 6 ml per kilogram has been shown to improve survival and decrease the number of days of ventilatory support in patients with the acute respiratory distress syndrome.1
Although it has not been proved that a temporary change in the tidal volume, to 10 ml per kilogram, may be deleterious, it is disturbing that the researchers, as well as their institutional review board, did not consider such a change important enough to require informed consent. It is an axiom of good research and protection of patients' rights that even an apparently trivial and noninvasive measurement should be explained to the patient, a surrogate, or both. Such measurements should be clearly outlined in the informed-consent document. The authors' suggestion that noninvasive measurement of the dead-space fraction is routinely used for nutritional assessment does not relieve them of their obligation to obtain informed consent when the measurement is performed for research purposes. Furthermore, when the dead-space fraction is measured for the purpose of nutritional assessment, it is not necessary to measure it routinely at a tidal volume of 10 ml per kilogram.
1 ReferencesTaek Sang Yoon, M.D.
Yizhak Kupfer, M.D.
Sidney Tessler, M.D.
Maimonides Medical Center, Brooklyn, NY 11219
stessler@maimonidesmed.org 1
The Acute Respiratory Distress Syndrome Network
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: As several of the letters suggest, the mechanisms that account for the increase in VD/VT early in the course of the acute respiratory distress syndrome need to be more thoroughly determined. We could not assess the contribution of cardiac output and intrapulmonary shunt because it was not feasible to insert pulmonary-artery catheters in the 179 patients in our study. Future studies should more clearly define the relation between VD/VT and cardiac output, shunt, and perhaps most important, pulmonary vascular resistance early in the course of the acute respiratory distress syndrome.
Patel and colleagues cite a recent study that demonstrated the value of volume resuscitation in patients with sepsis,1 not the acute respiratory distress syndrome. The value of an increase in cardiac output in patients with the acute respiratory distress syndrome is not known. The Acute Respiratory Distress Syndrome Clinical Network of the National Institutes of Health is carrying out a large, prospective, multicenter trial to evaluate a conservative fluid strategy as compared with a liberal fluid strategy, guided by hemodynamic measurements with either a pulmonary-artery catheter or a central venous catheter, in patients with acute lung injury (http://www.ardsnet.org). The results of this study should provide new information about the relations among cardiac output, intravascular volume, and mortality in patients with the acute respiratory distress syndrome.
Correction for compressible volume of the ventilator circuit was performed with the use of established methods in all patients, not just some patients, as Her suggests. We repeated the statistical analysis using corrected values for all patients; as we reported, the results remained significant. Previous studies, including one by Forbat and Her,2 have shown that measurement of VD/VT without the Frumin valve is accurate. Furthermore, Forbat and Her suggested that the Frumin valve may cause expiratory obstruction and compromised cardiac output when positive end-expiratory pressure is applied.2 Although intrapulmonary shunt was not measured, the ratio of PaO2 to FiO2 was entered into all multiple logistic-regression models, and the association between VD/VT and mortality was independent of the degree of hypoxemia.
Measurements of expired carbon dioxide are safe, routine, and noninvasive, and the requirement for informed consent was waived. All patients were studied before publication of the results of the low-tidal-volume trial of the National Institutes of Health,3 at a time when there was no clear evidence that lower tidal volumes were efficacious. In fact, one group had reported that ventilation with a lower tidal volume (7 ml per kilogram) was not beneficial.4 Thus, at the time of our study, 10 ml per kilogram was considered to be an intermediate and safe tidal volume for patients with the acute respiratory distress syndrome.
We chose to report odds ratios, which is standard practice for reporting the results of logistic-regression analyses. The formula used by Dr. Ho to calculate the relative risk5 was designed for use with dichotomous predictor variables, not with continuous predictor variables such as VD/VT .
5 ReferencesThomas J. Nuckton, M.D.
Mark D. Eisner, M.D., M.P.H.
Michael A. Matthay, M.D.
University of California, San Francisco, San Francisco, CA 94143-0130
tomnuc@itsa.ucsf. edu 1
Rivers E ,Nguyen B ,Havstad S , et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368-1377
Full Text | Web of Science | Medline2
Forbat AF ,Her C . Correction for gas compression in mechanical ventilators. Anesth Analg 1980;59:488-493
CrossRef | Web of Science | Medline3
The Acute Respiratory Distress Syndrome Network
Full Text | Web of Science | Medline4
Stewart TE ,Meade MO ,Cook DJ , et al. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. N Engl J Med 1998;338:355-361
Full Text | Web of Science | Medline5
Zhang J ,Yu KF . What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998;280:1690-1691
CrossRef | Web of Science | Medline
