To determine whether responses to dehydration are altered with age, we investigated the thirst, fluid and electrolyte responses, and hormonal responses to 24 hours of water deprivation in seven healthy active elderly men (67 to 75 years old) and seven healthy young men (20 to 31 years old) who were matched for weight loss during water deprivation. After water deprivation, the older men had greater increases in plasma osmolality, sodium concentration, and vasopressin levels. However, their urinary osmolality was lower and they were less thirsty and drank less after water deprivation, so that their plasma and urine were not diluted to predeprivation levels. Regression analysis indicated increased sensitivity of vasopressin osmoreceptors in the older group, although this difference was not statistically significant. We conclude that after 24 hours of water deprivation, there is a deficit in thirst and water intake in healthy elderly men, as compared with younger men, although vasopressin osmoreceptor responsiveness is maintained or even increased. Our findings also suggest that the well-known deficit in urinary concentrating ability that occurs with age reflects renal causes and not a lack of circulating vasopressin.

We and others have previously demonstrated excessive phosphocreatine (PCr) depletion and acidosis in skeletal muscle during exercise in patients with congestive heart failure (CHF). In the present study, we performed serial measurements of PCr and pH during gradually incremental flexor digitorum superficialis exercise in 22 patients with CHF and 11 age-matched controls to determine: (1) whether abnormalities were present at the same relative workloads (a comparison that would at least partially compensate for differences in muscle mass), (2) the temporable course of the metabolic changes, (3) the relationship of the metabolic findings to clinical variables, and (4) the relationship of the metabolic abnormalities to forearm blood flow. The patients with CHF had significantly lower [PCr] and pH at all submaximal levels of exercise, and these abnormalities were apparent from the onset of low-level exercise. There was considerable heterogeneity among the patients with CHF with respect to the metabolic findings, with 14 of 22 exhibiting either PCr or pH values more than 2 SDs below normal. Patients whose capacity was more limited during the protocol had lower [PCr], and especially pH, at low loads than did other patients with CHF or the control subjects. The more symptomatic patients and those with more limited bicycle exercise tolerance also had lower pH values. In contrast, there were no significant differences in forearm blood flow between the patients and controls and no relationship between forearm blood and either clinical variables or the metabolic findings. These results indicate that skeletal muscle metabolic abnormalities are present in many patients with CHF and that they are not primarily due to either muscle atrophy or impaired blood flow. These changes may explain in part the marked heterogeneity of symptom status and exercise capacity of patients with similar degrees of cardiac dysfunction.

Previous studies with 31P nuclear magnetic resonance have demonstrated that patients with chronic congestive heart failure often exhibit increased glycolytic metabolism and impaired oxidative phosphorylation in exercising skeletal muscle, but the mechanism for these changes remains unresolved. This study was conducted to determine whether these abnormalities result from impaired blood flow or oxygen delivery. Nine patients with mild-to-moderate congestive heart failure and nine age-and size-matched, healthy control volunteers were studied during repetitive submaximal finger flexion exercise under aerobic and ischemic conditions. Skeletal muscle metabolism was assessed by 31P nuclear magnetic resonance of the flexor digitorum superficialis muscle. During steady-state aerobic exercise at 33% of each subject's predetermined maximum workload, the patients with congestive heart failure exhibited significantly lower pH values (6.65 0.22 vs. 6.97+± 0.09, p<0.002) and phosphocreatine concentrations, 0.79 + 0.08, p<0.002). Similar differences were also present throughout ischemic exercise at the same workload. Based upon these measurements, calculated lactate production and adenosine 5'-triphosphate consumption rates were significantly higher in the patients with congestive heart failure. These results indicate that in many patients with congestive heart failure exercising muscle exhibits increased glycolytic metabolism and appears to be metabolically less efficient in relation to external work performed. These changes cannot be explained by impaired blood flow or oxygen delivery alone. (Circulation 1988;78:320-326) E xercise intolerance is the cardinal symptom of chronic congestive heart failure (CHF), yet its severity correlates poorly with hemodynamic indexes of cardiac function.' 2 A growing number of studies with 31p nuclear magnetic resonance (NMR) by US3,4 and others5.6 has demon-