Recurrent pneumomediastinum in a patient with rheumatoid arthritis.

Spontaneous pneumomediastinum (SPM); also known as mediastinal emphysema, is a rare and usually benign self-resolving appearance of extraluminal air in the mediastinum without any underlying trigger. This is an uncommon disorder mostly seen in the young males and classic clinical presentation is with chest pain, dyspnea, cough and appearance of subcutaneous emphysema. Although several connective tissue disorders have been reported in association with SPM, it is a rare occurrence in rheumatoid arthritis (RA) with only small number of cases reported in literature. We report a 69 years old male with RA who developed recurrent asymptomatic episodes of SPM detected over a period of one year. The recurrent but benign episodes of SPM in this patient reestablish the usual uncomplicated course of this unusual clinical entity even in the rare recurrent cases.

Effect of opposite changes in cardiac output and arterial PO2 on the relationship between mixed venous PO2 and oxygen transport.

We examined the relationship between changes in systemic oxygen transport (SO2T) and mixed venous PO2 (PvO2) in nine critically ill patients with acute respiratory failure and analyzed the effect of like and opposite changes in cardiac output (CO) and arterial PO2 (PaO2) on this relationship. Paired measurements of oxygen consumption (VO2), SO2T, and PvO2 were obtained before and after changes in the level of positive end-expiratory pressure (PEEP) equal to or more than 5 cm H2O. VO2 was measured with a rebreathing circuit adapted to a volume ventilator, and SO2T was calculated from thermodilution CO, PaO2, SaO2, and hemoglobin. In eight studies, CO and PaO2 changed in the same direction, and the absolute change in SO2T averaged 48 +/- 38 ml/min/m2. In 12 studies, CO and PaO2 changed in opposite directions, and the absolute change in SO2T averaged 78 +/- 69 ml/min/m2. When PaO2 and CO changed in the same direction, PvO2 increased on the higher level of SO2T (average difference 3.0 +/- 3.7 mm Hg, p less than 0.05) and there was a strong positive correlation between the difference in SO2T on lower and higher levels of PEEP and the difference in PvO2 (r = 0.83). When PaO2 and CO changed in opposite directions, PvO2 was unchanged on the higher level of SO2T, and there was no correlation between the difference in SO2T on lower and higher levels of PEEP and the difference in PvO2 (r = -0.45). VO2 was not different at the lower and higher levels of SO2T in both groups, indicating that VO2 was not transport-limited in these patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Dual-isotope measurement of lung water.

Iodine-131-labeled iodo-antipyrine and 99mTc-labeled erythrocytes were used to measure water content in lungs. These radioactive tracers were injected into 11 dogs with injured lungs. Blood samples were drawn and the animals sacrificed. The lungs were removed, weighed and homogenized. Samples of blood and lung homogenate were assayed for 131I and 99mTc. Samples were also weighed before and after drying to a constant weight at 70-76 degrees C. Extravascular lung water was determined by the dual-isotope technique and again by gravimetric analysis. The average ratio of the results from the two different methods was 1.14 +/- 0.20. The two methods were also compared by regression analysis and the correlation coefficient was 0.97 +/- 0.09.

Perfusion distribution and lung thermal volume in canine hydrochloric acid aspiration.

We investigated the effects of a brief period of positive end-expiratory pressure (PEEP) ventilation or nitroglycerin (NTG) infusion on the distribution of pulmonary blood flow and extravascular thermal volume (ETV) in anesthetized dogs with unilateral HCl lung injury. ETV was determined by the thermal dye technique by use of a monoexponential extrapolation to exclude recirculating indicator, and regional blood flow was determined by a particle distribution technique (radiolabeled plastic microspheres). The lungs were weighted after the animals were killed, and extravascular lung mass (ELM) was determined with the use of hemoglobin to correct for trapped lung blood. Measurements were obtained before instillation of HCl into the right lung and repeated 3 h later before, during, and after PEEP ventilation or NTG infusion. Fractional perfusion of the severely injured portion of the right lung (Qinj/QT) fell from 44.3 +/- 11.1% at base line to 27.8 +/- 15.4% after the onset of lung injury. PEEP produced an acute reversible increase in ETV (63 +/- 37% over average of pre- and post-PEEP values), and the changes in ETV were closely correlated with changes in Qinj/QT (r = 0.91). NTG infusion produced insignificant increases in ETV (14 +/- 10% over average of pre- and postinfusion values) and Qinj/QT (59 +/- 35%), but the changes in ETV and Qinj/QT were strongly correlated (r = 0.92). The fraction of extravascular lung mass detected by the thermodilution measurement averaged 0.44 (range 0.24-0.77).(ABSTRACT TRUNCATED AT 250 WORDS)

Lung water measurements with iodo-antipyrine.

131I labeled iodo-antipyrine and 99mTc labeled erythrocytes were used to measure water content in lungs. These radioactive tracers were injected into 10 rabbits with normal lungs and 11 rabbits with injured lungs. Blood samples were drawn and the subjects were killed. The lungs were removed, weighed and homogenized. Samples of blood and lung homogenate were assayed for 131I and 99mTc. Samples were also weighed before and after drying to a constant weight at 70-75 degrees C. Extravascular lung water was determined by the dual isotope technique and again by gravimetric analysis. The average ratio of the results from the 2 different methods is 1.03 +/- 0.15. The 2 methods were compared by regression analysis and the correlation coefficient was 0.92 +/- 0.09. Our investigation suggests the possibility of measurement of lung water with equilibrium distribution of iodo-antipyrine.

Relationship between CO and transit times for dye and thermal indicators in central circulation.

We investigated two factors that may influence the estimation of lung water by the thermal-dye double-indicator-dilution method: 1) changes in cardiac output (CO), and 2) thermal equilibration with cardiac tissue. In theory, the difference between mean transit times of thermal and dye indicators (delta MTT) is proportional to the extravascular volume of distribution of the thermal indicator (VODev) and inversely related to CO. The delta MTT also includes a time element DT due to the difference in response times of the measuring instruments such that delta MTT = VODev/CO + DT. In nine anesthetized dogs we recorded 286 aortic thermal and dye curves following left atrial (LA) and right atrial (RA) injections as CO was increased from 2.35 to 6.65 ml X s-1 X kg-1 by isoproterenol infusion, and a regression of delta MTT on CO-1 was performed. DT was measured in vitro for comparison with the y-intercept. In six of nine dogs the slope of the regression for LA injections was not different from zero, indicating that there is no measurable volume of distribution for thermal indicator in cardiac tissue. For RA injections the relationship between delta MTT and CO-1 was linear in all experiments, with an average correlation coefficient of 0.97 +/- 0.01 (SE), indicating that the VODev was constant over a threefold increase in CO. Although the in vitro measurement of DT agreed closely with the average of the y-intercepts of the regressions, small between-subject differences in DT can lead to apparent flow-related changes in extravascular thermal volume computed in the conventional fashion using the in vitro estimate of DT.

Effect of PEEP and type of injury on thermal-dye estimation of pulmonary edema.

We investigated the effect of positive end-expiratory pressure (PEEP) on the extravascular thermal volume of the lung (ETV) determined by the thermal-dye technique in three canine models of pulmonary edema created by injection of alpha-naphthylthiourea (ANTU) or oleic acid (OA) into the pulmonary circulation or intrabronchial instillation of hydrochloric acid (HCl). ETV was determined before, during, and after ventilation with 14 cmH2O PEEP, and final ETV was compared with the extravascular lung mass (ELM) determined postmortem. Final ETV correctly estimated ELM in 12 animals with ANTU injury, ETV/ELM = 1.04 +/- 0.13, but underestimated after HCl injury (n = 5), ETV/ELM = 0.61 +/- 0.23, and OA injury (n = 6), ETV/ELM = 0.73 +/- 0.19. Whereas PEEP had no consistent effect on extravascular thermal volume in ANTU edema, there was a reversible increase in ETV during PEEP in animals with HCl or OA injury and underestimation of ELM. The increase in ETV during PEEP averaged 9.3 +/- 3.8 ml/kg (62 +/- 42%) over the mean of the pre- and post-PEEP values after HCl injury (P less than 0.01) and 6.7 +/- 4.4 ml/kg (47 +/- 35%) after OA injury (P less than 0.02). There was an inverse correlation between the change in ETV during PEEP and the ETV/ELM ratio for animals with HCl and OA injury (r = -0.94). We conclude that PEEP produces a reversible increase in ETV in some models of lung injury by allowing for distribution of thermal indicator through a larger fraction of the lung water and that this response may be useful to detect underestimation when gravimetric measurements are not available.

Thermodilution measurement of lung water.

The detection and measurement of pulmonary edema by the thermal-dye method appears to be accurate and reproducible under specified laboratory conditions. The ETV, which represents the difference in distribution volumes of the diffusible (thermal) indicator and the intravascular (green dye) indicator, should closely estimate the ELM (ETV = 0.984 ELM). Experimental measurements of ETV have shown a very good correlation with ELM, with a tendency for overestimation in normal lungs and underestimation in severely edematous lungs. In contrast to previous measurements using isotopic water methods, thermal-dye measurements have revealed that the estimation of ELM by ETV in severe edema (alveolar flooding) does not plateau. The limitations of the thermal-dye technique reflect the evenness of lung perfusion. Depending on their size and number, emboli produce perfusion defects and reduce ETV. Airway injury also reduces ETV, apparently by redistribution of blood flow. Alterations of ETV by hemodynamic factors suggest that reduction in perfusion pressure may be more significant than changes in flow, although more data are needed. Atelectasis without a reduction in blood flow does not decrease ETV. PEEP may increase ETV when lung injury is not uniform, perhaps by redistributing blood flow, and this maneuver may be useful in detecting underestimation of ELM. Position of the thermistor produces the greatest degree of variability by distorting the thermodilution curve and prolonging the MTT. This results in an increased ETV and an overestimation of ELM. In laboratory studies, the measurements of ETV can be validated by gravimetric analyses of lung water. Since this method of validation is not possible in clinical studies, measurements of ETV in patients must be interpreted in light of limitations demonstrated in the laboratory. Suggestions for avoiding the most common errors in measuring ETV are listed in Table 3.

Type of lung injury influences the thermal-dye estimation of extravascular lung water.

To determine the effect of the type of lung injury on the thermodilution estimation of extravascular lung water, we produced pulmonary edema in 25 anesthetized dogs by injection of alloxan or alpha-naphthylthiourea (ANTU) into the pulmonary circulation or by instillation of hydrochloric acid (HCI) into the airway. HCl injury was bilateral, unilateral with tidal volume equal in each lung, or unilateral with equal airway pressure. Extravascular thermal volume (ETV) was measured at base line and 4 h after lung injury, and the final measurement was compared with the postmortem determination of extravascular lung mass (ELM). In 11 of 15 animals with HCl injury final ETV was less than the base-line measurement. The ratio of final ETV to ELM for all HCl animal (group I) averaged 0.31 +/- 0.14, which was different from the value for animals with alloxan or ANTU injury (group II), 1.04 +/- 0.14 (P less than 0.01). Extravascular lung water per gram of blood-free dry tissue was not different for the two groups (8.1 +/- 1.2 and 8.7 +/- 2.6 for I and II, respectively), indicating equally severe lung injury; however, shunt fraction was less in group I (P less than 0.01). ETV/ELM correlated with the shunt fraction for group I (r = 0.70) but not for group II (r = 0.32). These findings indicate that ETV underestimates lung water after HCl injury due to the redistribution of pulmonary blood flow away from edematous areas.

Effect of edema and hemodynamic changes on extravascular thermal volume of the lung.

The extravascular thermal volume of the lung (ETV) has been measured in dogs as the difference between mean transit time (t) volumes for heat and indocyanine green dye across the pulmonary circulation, calculated as the product of thermal dilution cardiac output (CO) and the difference in t for aortic indicator-dilution curves generated by right and left atrial injections. ETV measurements were compared with the extravascular lung mass (ELM): in 21 normal dogs, ETV/ELM = 1.11 +/- 0.14 (SD); in 17 dogs with hydrostatic pulmonary edema (up to 21 g/kg), ETV/ELM = 0.90 +/- 0.11; and in 27 dogs with alloxan pulmonary edema (up to 51 g/kg); ETV/ELM = 0.93 +/- 0.13. For all 65 dogs the mean ETVELM was 0.98 +/- 0.15, and the liner regression was ETV (ml/kg) = 0.90 ELM (g/kg) + 0.86 +/- 2.25 (SEE; r = 0.96). Calculations based on measurements of lung specific heat predict that ETV/ELM should equal 0.984. With acute changes in pulmonary hemodynamics, ETV was reduced by reductions in pulmonary arterial pressure (Ppa) sufficient to produce zone 1 conditions at the top of the lung. However, ETV was not affected by increases in CO (mean = 50%) produced by nitroprusside or by increases in Ppa and pulmonary blood volume (mean = 27%) produced by partial mitral valve obstruction. Distortion of the thermal dilution curve due to position of the arterial thermistor appears to be the greatest source of variability and overestimation. Simultaneous measurements from pairs of thermistors differed by 14% (range 0.4-50%).