Published ahead of print on September 11, 2003, doi:10.1164/rccm.200307-1016OE
© 2003 American Thoracic Society
Bill Dock and the Location of Pulmonary TuberculosisHow Bed Rest Might Have Helped ConsumptionUniversity of California, San Francisco, California Correspondence and requests for reprints should be addressed to John F. Murray, M.D., Box 0841, Pulmonary and Critical Care, University of California, San Francisco, CA 94143–0841. E-mail: johnfmurr{at}aol.com The idea that sick lungs would benefit from a period of enforced rest is very old and sounds compelling, even though during the 100 or so years that physicians did their best to impose it, there was a total lack of clinical and physiologic evidence to support the practice. The belief that broken and sore lungs required rest was one of the first things the writer Betty MacDonald (1), newly ill with tuberculosis, discovered when she was sequestered in a sanatorium called The Pines: "If you had a broken leg you wouldn't dance on it nor walk on it but would have a plaster cast or splints on it so that you couldn't use it even if you were foolish enough to try. If you had a sore on a joint or a knuckle, you would know that constant bending would break the sore open and prevent its healing quickly. When you have tuberculosis you have broken lungs with sores on them and the less you use them the quicker they will heal. How can you rest your lungs? By breathing less often and less deeply. A person resting quietly in bed, breathes two times less each minute than a person sitting up and of course much less than a person walking... Rest is the answer. Rest, rest and more rest." Lying in bed, as Betty MacDonald and hundreds of thousands of other tuberculars did for months on end, is an easy way of resting the lungs, but slowing breathing a couple of breaths each minute is not nearly as efficient as putting lungs completely to sleep. That goal was aggressively sought after by phthisiologists and surgeons who came up with pneumothorax, pneumoperitonium, and thoracoplasty, which were mainstays of the treatment of tuberculosis for over half a century. The rationale for immobilization was advanced by Carlo Forlanini (2) in 1882, the same year that Robert Koch announced his discovery of Mycobacterium tuberculosis; Forlanini wrote that the cure of pulmonary tuberculosis was impossible because the lungs were "in an unceasing motion of expansion and reduction. When this peculiarity is removed, the [static] lung becomes similar to other viscera and from that moment [healing begins]." This sounds good, but in spite of pronouncements of this sort, perhaps even because of them, the purported clinical benefits of bed rest and collapse therapy were never properly evaluated. The purpose of this review is to re-examine the physiologic evidence pertaining to these treatments and to emphasize why we should have listened to one man, William (Bill) Dock, who told us all long ago how bed rest should be performed and why certain types of collapse might be helpful. BACKGROUND One of the remarkable features of consumption, which had piqued the curiosity of early students of the disease, was its distinctive distribution within the lungs. Valsalva may have been the earliest person to point out that the lesions of reactivation pulmonary tuberculosis in humans are often confined to the upper lobes, and Morgagni sharpened the location to the upper parts (the apices) of the upper lobes. These observations were reinforced 65 years later by Laënnec, who also noted that early cavities seemed to favor the right side over the left, as shown in Figure 1 (3).
More recent confirmation comes from Medlar's search for the intrapulmonary locations of what used to be called "minimal" tuberculosis (i.e., small noncavitary lesions) (4). During his dissections of 1,225 victims of sudden or violent death, he found 96 lesions of active minimal disease: There was a striking apical–posterior segmental predilection and a right-sided predominance of lesions in the upper third of the lungs. When four-legged animals develop pulmonary tuberculosis, either acquired naturally or induced experimentally, the lesions concentrate in the dorsal regions of the lungs, those zones that are just beneath the ribs in the back and that are uppermost when the animals are on their feet; in contrast, when four-legged animals are held in the erect position by a harness for 11 hours a day, the localization of experimentally induced tuberculosis shifts from the dorsal regions to the apices, similar to that in humans. Bats provide another variation on the positional theme; when they develop tuberculosis, the lesions occur in the bases of their lungs, the zones that are uppermost during the daytime while the animal is hanging upside down sound asleep (see Reference 3 for details and citations). What could possibly explain these seemingly bizarre preferences? This is an interesting question—one with therapeutic implications but one that had baffled the experts for 2.5 centuries. The answer was finally provided by Bill Dock, an extraordinary man, who came up with a very simple explanation for the apparently freakish anatomic distribution of reactivation pulmonary tuberculosis: gravity, which has important effects on both the distribution of inspired air and the distribution of pulmonary arterial blood flow within the lungs. The former determines the sites of deposition of inhaled M. tuberculosis and, hence, the location of the initial lesions of tuberculous infection. The latter creates a regional environment within the lungs that favors the selective growth of tubercle bacilli implanted during the early phase of widespread hematogenous dissemination, thereby determining the characteristic sites of reactivation (or postprimary) lesions (5). INITIAL LESIONS During inhalation, fresh air entering the lungs is preferentially directed toward their lowermost portions in whatever body position we happen to be in: to the bases when we are sitting or standing or to the dorsal regions when we are lying on our backs. To direct an entering breath to the apices, we would have to stand on our heads. Inspired air always heads downward to whatever zones are most dependent. The bottom-seeking distribution of inhaled air occurs because of gravity, which gives the lungs weight and affects the way they hang in the chest cavities. Simple but elegant experiments performed in the breathing laboratory of the NASA space shuttle confirm, as would be expected, that when weightless, the gravity-dependent effects observed on earth disappear, but some unevenness in the distribution of inspired gas remains (6). Because tuberculosis is transmitted as an airborne disease, the tiny infectious droplet nuclei, like the inspired air they travel with, are preferentially distributed to the dependent zones of the lungs. Thus, the initial lesions of tuberculosis, the Ghon focus, which develop at the site(s) of implantation of M. tuberculosis, tend to occur in the lower half of the upright lungs; this means that most exposures take place while the new victim is up and around and not while lying down. There is another interesting feature about the primary lesions of tuberculosis that has been carefully documented by pathologists in the past but is seldom commented on today: their location. In his compendium, The Behavior of Pulmonary Tuberculous Lesions, Medlar (7) wrote, "Usually the [initial] parenchymal lesions were found within 1 cm. of the pleura in the collapsed lung." Two other giants of tuberculosis pathology, Auerbach (8) and Pinner (9), located the primary lesions "just beneath the pleura" and "in the subpleural layers," respectively. The concentration of primary tubercles near the visceral pleural surface does not occur by accident. It means that the dynamics of airflow within the bronchial tree favor distribution of incoming droplets to the outermost regions of the dependent lungs, presumably because of the branching pattern of airways. At each of the roughly 25 branches of the bronchial system that conduct gas to the distal regions of the lungs, the branch serving the most peripheral pathway is straighter and has a larger cross-sectional area than the smaller, more sharply angled branch leading to proximal gas exchange units (10); inhaled droplet nuclei are more likely to escape being deposited on airways and to reach the alveolar surface if they are carried along the axial airstream within the straighter and larger of the two possible pathways afforded at each branch (Figure 2) .
REACTIVATION LESIONS The gravity-dependent distribution of inspired air and the pattern of airflow within the branching airways provide a straightforward explanation for the subpleural lower zone preference of the Ghon focus, but they do not explain the upper zone predominance of lesions in the even more common reactivation variety. That is what Bill Dock figured out. During the early 1940s, as part of their war-related studies to learn more about the circulation and the effects of injuries on blood pressure and cardiac output, André Cournand and Dickinson Richards resurrected an old technique performed by a young German surgeon named Werner Forssman, who while watching under a fluoroscope, performed the world's first cardiac catheterization with a long thin rubber tube that he passed through a vein in his arm into his own heart (Forssman, who shared the 1956 Nobel Prize with Cournand and Richards, performed this feat in 1929 but was then squelched by his superiors in the authoritarian German academic system from ever doing it again [11]). Members of the Cournand-Richards research team repeated Forssman's experiment and had no trouble catheterizing the right atrium and right ventricle, where they measured the blood pressure; they did not catheterize the pulmonary arteries in those early studies, but they inferred what the systolic pressure in the vessels had to be from the corresponding pressure in the right ventricle (12). In a later more detailed study, the blood pressure perfusing the lungs of healthy subjects proved to be quite low, only about one-fifth of the normal systemic arterial blood pressure (13). Dock knew at once what this observation meant: Because the arterial pressure in the lungs was so low, there had to be profound unevenness in the quantity of blood flowing to the highest compared with the lowest zones. Those regions that are uppermost—the apices in erect humans, the dorsal areas in upright four-legged animals, and the bases in upside-down bats—would be relatively starved of blood flow because the prevailing blood pressure in those locations was extremely low. Originally, Dock (5) postulated that this gravity-dependent phenomenon would create a critical difference in how the upper and lower zones of the lungs would be able to withstand infection and ongoing onslaught by M. tuberculosis. Because the upper regions had so little blood flow, they were robbed of protective circulating constituents such as phagocytes and antibodies, and formation and flow of beneficial lymph, which requires an intravascular hydrostatic pressure higher than intravascular colloid osmotic pressure, would be virtually nil. Without these growth restraints, multiplication of bacilli within the tiny foci of tuberculosis that were seeded in the upper zones during the early phase of widespread bloodstream dissemination would be greatly enhanced; in contrast, the lower zones, where blood flow was plentiful, were well supplied with defensive ammunition and abundant lymph drainage. Dock reinforced his belief in the importance of pulmonary arterial pressure and blood flow in the development of reinfection tuberculosis by citing some impressive information about its prevalence in various types of heart disease. In mitral stenosis, for example, in which pulmonary arterial pressure is extremely high and the lungs are usually congested, pulmonary tuberculosis is practically unknown, despite the patient's frequent wasting and general debility. In contrast, in patients with stenosis of the right ventricular outflow tract in whom pulmonary arterial pressure is low and the lungs are ischemic, consumption is extraordinarily frequent (14). A few years after his first report, Dock (3) updated his theory to take into account new information about how the topographic differences in blood flow and ventilation, both of which are influenced by gravity but blood flow far more than ventilation, affected the values of oxygen and carbon dioxide within the air spaces and blood vessels; he added the important fact that M. tuberculosis, which flourishes in an environment of high oxygen, would find "favorable soil" in the upper lung regions, where the concentration of oxygen is highest. Twenty years later, direct measurements of the distributions of blood flow and ventilation in the human lung using radioactive tracer gases and the resulting calculations revealed that the partial pressure of oxygen in the apex of the lungs (132 mm Hg) is approximately 50% higher than that in the lung bases (89 mm Hg), thus supporting Dock's theory (15). In his original article on the apical location of tuberculosis, Dock (5) also provided an explanation for the observations of Laënnec and Medlar that tuberculosis lesions were usually more extensive in the right lung compared with the left. Dock postulated that because of anatomic differences, particularly the sharply angled turn that blood flowing to the right upper lobe must make (the pathway is much straighter on the left), the blood pressure "must be somewhat lower in the [uppermost] branches of the right pulmonary artery than in the same branches of the left" (Figure 3) . Because blood pressure determines blood flow, assuming vascular resistance is the same in corresponding vessels, perfusion to the right upper zone must be less than that to the left, thus accounting for the difference in the extent of tuberculosis on the two sides. Many years afterward, quantification of intrapulmonary blood flow to the two lungs of healthy humans validated Dock's conclusion (16). The investigators who made the measurements end their confirmatory article as follows: "It is interesting that the predictions that Dock made about the apical pulmonary blood flow, from the localization of pulmonary tuberculosis in patients with a normal cardiovascular system, are exactly those found experimentally in [our] study."
BED REST Dock's physiologic reasoning provided the first plausible explanation of why bed rest, a recommended treatment for tuberculosis for over a century, might actually be beneficial. Not any kind of bed rest—it had to be in the recumbent position. To increase blood flow to the upper regions and maintain their lymph flow and oxygen tension equal to those in the lower regions, patients had to be flat in bed, either supine or prone. Recumbency was, in fact, rigorously enforced in some tuberculosis sanatoriums in which patients were fed, made to use bed pans, and bathed, all while lying flat. Dock would have approved, but even he recommended "brief periods of erect posture, which make eating, bathing and elimination more agreeable" and which improve morale and compliance "for the long periods of time which experience has proved necessary." For centuries, tuberculosis was considered an incurable disease; an occasional spontaneous remission, perhaps, but no treatment regimen consistently benefited. In 1853, when a German, Hermann Brehmer, claimed he could cure consumption with a regimen consisting of bed rest and restricted exercise in fresh air, most people did not believe him, and one expert called him a charlatan. Peter Dettweiler, originally a patient and later assistant to Brehmer, was the first person to use and promote prolonged rest—in the recumbent position and in bed when patients were febrile and in comfortable couches outdoors or in sheltered porches when afebrile. In 1886, Dettweiler reported that 72 patients out of more than 1,000 had remained well for longer than 3 years—better than usual, apparently, but not a brilliant success rate (17). Edward L. Trudeau was an early American advocate of cure by prolonged bed rest, but he took his share of abuse about it too. Most authorities did actually practice bed rest at the beginning of treatment, particularly while patients were febrile. At issue was whether bed rest should be enforced for long periods as Dettweiler, Trudeau, and Joseph Pratt preached or whether it should be supplemented early on by graded exercise as other experts decreed. In his review of the debate, Pratt (17) claimed a statistical advantage to prolonged bed rest over gradual exercise, but few numbers are provided. There are no p values, and satisfactory control subjects were never included. In fact, no randomized prospective controlled trials were ever performed to evaluate any form of bed rest, including Dock's version using recumbency. COLLAPSE THERAPY Dock (3) also applied his mastery of hemodynamics and gas exchange to the mechanisms by which pneumothorax, another popular form of therapy for consumption that was never rigorously evaluated, might also be beneficial. Total collapse of a lung or lobe reduces blood flow substantially and arrests ventilation. Consequently, the oxygen tension in the collapsed lung parenchyma reaches its lowest possible value, equal to that in mixed venous blood (approximately 40 mm Hg). Pneumothorax may have helped pulmonary tuberculosis in ways other than by decreasing oxygen tension, but interrupting the "unceasing motion" of the lungs may not necessarily be one of them. Dock cited the experiments of Rich and Follis (18), who showed that tuberculosis in guinea pigs could be greatly attenuated when the animals were given 10% oxygen to breathe, despite the resulting vigorous hypoxia-induced hyperventilation. In this model at least, reduced alveolar oxygen tension overcame any adverse effect of exaggerated breathing movements on the growth of tubercle bacilli. DENOUEMENT The arguments over the putative benefits of prolonged bed rest were muted by the arrival of isoniazid in 1952 and were effectively quashed in 1959 when Wallace Fox and his British Medical Research Council colleagues (19) proved that bed rest and sanatorium treatment plus antituberculosis chemotherapy were no better than the same chemotherapy administered to fully ambulatory patients at home, provided the patients received the right combination of medications and took them. Collapse therapy also was soon rendered obsolete by effective antibiotics. However, Dock (3) had one last suggestion. In 1954, he wrote, "Strains of bacilli resistant to known and yet-to-be discovered antibiotic combinations may appear, but strains which are insensitive to the tension of gases in venous blood are not known." He seemed to be prophesizing the arrival of multiple drug-resistant tuberculosis, which is difficult and costly to treat in low-income countries and for which bed rest might be beneficial until the necessary drugs and facilities become more widely available. FOOTNOTES Conflict of Interest Statement: J.F.M. has no declared conflict of interest. Received in original form July 23, 2003; accepted in final form September 4, 2003 REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
