Certain characteristics of dyspnea suggested that this symptom might be explored by the same means employed in studying pain associated with pathological tissue changes. A regular pattern is observed in certain types of dyspnea related to the time of day. Cardiac asthma or paroxysmal cardiac dyspnea occur mainly at night after patients have gone to sleep. However, this is not due to the recumbent position since patients may assume this position during the day without any ill effect. Harrison's "evening dyspnea" (20) is characteristically absent in the morning, but develops slowly during the course of the day, reaching a maximum in the evening.

Just as for pain, the degree of intensity of dyspnea was studied in relation to hourly acid base balance changes as indicated by changes in the pH of the urine. Patients having dyspnea of prolonged duration as a result of various pathological conditions, with no treatment of any kind for at least six hours before or during the period of observation, were the subjects of this investigation. The degree of intensity of the dyspnea was estimated by trained observers who were in constant attendance. Dyspnea was recorded by the observers as absent, slight, moderate, severe, or very severe, estimations being based upon rate depth and evident difficulty in respiration.

At the time of these observations, hourly urine specimens were obtained from patients with as little disturbance as possible. No patient showed evidence of renal disease. The pH of the urine specimen was determined potentiometrically. The curves showing hourly fluctuations in the intensity of the dyspnea and the changes in the pH of the hourly urine specimens were then plotted and compared. Acidifying and alkalizing substances were administered to patients during the course of some tests in order to observe the influence of induced changes in the acid base balance upon the degree of dyspnea. Phosphoric acid and sodium bicarbonate were used for this purpose.

Fourteen patients with different pathological conditions were studied. In ten cases, a distinct correlation between the intensity of the symptom and acid base variations was found.

Four patients had pulmonary edema associated with the symptom of dyspnea. One patient had edema due to congestive heart failure; another had pulmonary edema and lung metastases from a carcinoma of the pancreas. Two other patients with cancer of the breast metastatic to the bone and skin also had pulmonary edema. In one of these, the pulmonary edema appeared to be a result of the accidental introduction of a fatty acid in oil preparation into the blood stream following an intramuscular injection. In all four of the cases with pulmonary edema, the intensity of dyspnea was found to be increased when the urinary pH showed changes toward more alkaline values and was diminished when the pH changes were opposite. In these cases, the intensity of the dyspnea was relieved following administration of phosphoric acid, while sodium bicarbonate increased the dyspnea. By analogy with pain, we called this correlation an alkaline pattern.

The parallel variations between the curves of the intensity of dyspnea

Fig. 34. The parallel variations between the curves of the intensity of dyspnea and of the value of the urinary pH in a case of pulmonary congestion following an accidental intravenous injection of a fatty acid preparation, indicates an alkaline pattern.

Six cases showed an opposite type of correlation between the intensity of dyspnea and the acid base changes in the pH of the urine. All of these cases had mediastinal or pulmonary masses and failed to show signs of pulmonary edema. In these cases, the maximum degree of dyspnea was associated with a relatively more acid urine, and the dyspnea was less intense when the urine was more alkaline. In these cases, phosphoric acid increased the degree of dyspnea, while conversely sodium bicarbonate decreased it. This would correspond to an acid pattern of dyspnea. (Figs. 35, 36)

These findings, although obtained in only a limited number of patients, strongly suggest a similarity between the fundamental origin of both pain and dyspnea. As in pain, the two patterns of dyspnea were associated with a relative alkalosis and a relative acidosis.

Fig. 35. The opposite concomitant variations between the intensity of dyspnea and urinary pH in a case of mediastinal metastases of a hypernephroma indicate an acid pattern.

The opposite concomitant variations between the intensity of dyspnea

Certain differences exist between the investigations of pain and dyspnea. In studying pain, it was necessary to depend entirely upon the observations of the patient as to the relative intensity of the pain experienced from hour to hour. In dyspnea, the patient's own observations were found to be less reliable due to the emotional factors associated with dyspnea. It was, however, possible to depend upon trained observers who could estimate accurately enough the degree of dyspnea on the basis of rate, depth and apparent difficulty of respiration. It was found that the greater the experience of the observer, the closer was the correlation between the curves of intensity of dyspnea and the changes of the urinary pH values.

The administration of sodium bicarbonate increases the intensity of the dyspnea

Fig. 36. The administration of sodium bicarbonate increases the intensity of the dyspnea in a case of pulmonary metastases of cancer of the gall bladder, indicating an alkaline pattern.

Since it appeared evident that acid base changes play a significant role in influencing the degree of dyspnea experienced by the patient in two opposite directions, it was important to consider this influence in relation to the physiology of respiration which is known not only to be affected by, but even dependent on acid base changes.

Concerning the respiratory center, two different mechanisms of direct chemical stimulation have been suggested as being responsible for dyspnea. One is lack of oxygen, but considerable evidence exists to indicate that simple anoxemia is not a true cause of dyspnea. (21, 22) On the other hand, an acidosis of the respiratory center is known to cause dyspnea, but it has been established that only a very profound general acidosis, such as that connected with diabetes, acid poisoning or emphysema can create a change in blood pH sufficient to affect the respiratory center. This is also true for the chemoreceptor centers located in the carotid sinus, which are still less sensitive to the general lack of oxygen or acidosis than the respiratory center. While general acid base changes seem to play an important role in dyspnea, these chemical changes would not appear to act direcdy upon the respiratory center since the actual blood pH is not sufficiendy changed.

Other factors—involving direct local, rather than systemic influences— may be considered. One such direct action would be related to reflex stimulation which is generally accepted as having an important role in the control of respiration and even dyspnea. This reflex control of respiration is connected with nerve endings within the lung parenchyma which are stimulated when the walls of the alveoli are stretched in the respiratory phase. The stretch reflex induces impulses, carried by way of the vagus, which acting upon the respiratory center, stop the inspiratory phase and bring about expiration. It may be assumed that the nerve endings within the parenchyma can be stimulated not only by mechanical changes within the parenchyma but also by local chemical changes too. Under abnormal conditions such as pulmonary congestion, there may be a change in the tissue reaction, as has been recognized by a higher pH of the edema fluid itself. Through this reflex mechanism, these pulmonary tissue reaction changes may produce dyspnea. The general acid base fluctuations influencing the local alkalosis would, as seen in pain, indirectly influence the degree of dyspnea. Such a mechanism may account for dyspnea with an alkaline pattern found in pulmonary congestion.

A similar local factor also could be seen for the acid pattern. It was observed that all paUents with an acid pattern of dyspnea had tumors located within the mediastinum or in the lung parenchyma itself. From the study of pain, we know that abnormal degrees of acidosis can occur in tumors. pH changes toward acidosis within a tumor tissue in the vicinity of chemoreceptors may produce impulse discharges in these centers especially sensitive to changes toward acidosis. They may alter the character of respiration and result in dyspnea. As in other conditions, these local changes occur even with reduced changes in the systemic acid base balance. Through this mechanism, variations in the intensity of dyspnea can occur without the intensive change in the general acid base balance which is considered necessary to affect directly the respiratory center itself.

The fact that relatively small acid base balance changes affect the intensity of dyspnea in two opposite directions can thus be explained by this indirect influence exerted upon a local process, leading to alkalosis in abnormal conditions affecting the pulmonary parenchyma and to acidosis in lesions present in the neighborhood of the chemoreceptive centers. These two mechanisms proposed as part of this working hypothesis, appear able to explain the paradoxical experimental findings in clinical studies of this symptom, where opposite responses upon dyspnea are seen for the same acidifying and alkalizing agents.

A more complete study of dyspnea under this aspect will be published separately.