Theophylline Therapy for Near-Fatal Cheyne-Stokes Respiration

IN RESPONSE:

Sin and Bradley and Geigel and Chediak point out that the respiratory tracings do not represent classic Cheyne-Stokes respiration (CSR). Qualitative monitoring of chest wall movements depends on the gain and nonlinear transfer characteristics of the respiratory monitor used and may not always reflect CSR. Part A of the Figure shows a respiratory tracing in of the chest movements in a patient having the typical pattern of CSR (1), evident on simultaneous, more sensitive measures of breathing. Compare this figure to one of those to which the correspondents object (Figure, part B). There is little difference between these two tracings. Other considerations are that breathing disorder patterns are not mutually exclusive but rather frequently coexist and that breathing patterns are labile and are easily affected by apprehension, laboratory environment, and instrumentation. Part C of the Figure shows a tracing from our index patient, which better reflects CSR. We chose the tracings not to illustrate CSR but rather to highlight other, more novel aspects of the report.

View larger version:

• In this window
• In a new window

Figure. In patient with Cheyne-Stokes respiration, as reported by Naughton et al. (1). The tracing is similar to the one in question from Pesek et al. ( ). A tracing from the index patient that is more representative of Cheyne-Stokes respiration. Chest wall movements. A.BC.

The observations by Cheyne in 1818 and Stokes in 1854 made no reference to CO₂ or other pathophysiologic features as pathognomonic characteristics. Although patients with congestive heart failure (CHF) and CSR often have transcutaneous PCO ₂ measurements indicating hypocapnia, Sin and Bradley overstate the universality of this observation and overlook important pathophysiologic information. Arterial blood gas measurements by Dowell and colleagues (2) showed peak CO₂ levels of 45 to 73 mm Hg in 4 of 10 patients with CSR (1 with lung disease). Massumi and Nutter (3) described peak CO₂ levels of 52 to 62 mm Hg during CSR in 3 of 7 patients with CSR who also had profound dysrhythmias unquestionably related to CSR (a clinical picture similar to our patient's).

We are perplexed at Sin and Bradley's assertions that respiratory stimulants are well-recognized for treating central alveolar hypoventilation syndrome and that theophylline's effectiveness in our patient is unsurprising. They cite Bradley and Phillipson, who stated that in their experience theophylline “does not provide long-term improvement in adults with hypercapnic CSA.” The absence of a ventilatory response to aminophylline has even been suggested as a possible aid to diagnosis of central alveolar hypoventilation syndrome (CAH) (4). Pulmonary texts that address theophylline therapy for CAH do so simply to dismiss theophylline as seldom beneficial in CAH. Our patient's swift, absolute, and long-term response to theophylline is not easily reconciled with a diagnosis of CAH.

Many studies have demonstrated CSR resolution with theophylline. With regard to use of theophylline for treating CSR in CHF, our report should have no direct relevance to changing patient care in general. Indeed, we emphasized that an interesting feature of our patient was the absence of CHF. Any potential effect of theophylline on CSR in CHF has been addressed more thoroughly elsewhere. However, theophylline at maximal therapeutic concentrations has minimal phosphodiesterase inhibitory activity (5). Thus, although we cannot speak to the long-term effects of theophylline for treating CSR in patients with CHF, Sin and Bradley's reasoning is inconsistent with current knowledge and is unlikely to be germane to the question.

• Copyright ©2004 by the American College of Physicians