INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

TYPE 2 DIABETES IS ASSOCIATED with obesity and sedentary living (17); however, maximal and submaximal O₂ consumption (O₂) are lower in patients with Type 2 diabetes than in nondiabetic individuals of similar age, gender, body composition, and self-reported physical activity (7, 8, 10, 19, 20, 27). These findings suggest that some characteristic of diabetes other than fitness or body composition contributes to decreases in maximal O₂ (O_2 max). With the exception of aerobically fit, healthy individuals who can be limited by pulmonary gas exchange (4), O_2 max is affected by limitations in cardiac output (), redistribution of blood flow to working muscle, and skeletal muscle O₂ extraction in healthy individuals (21). However, little is known about the mechanism responsible for the lower O_2 max in Type 2 diabetic patients than in similarly unfit nondiabetic subjects.

There is indirect evidence suggesting that exercise may be impaired by Type 2 diabetes. Up to 60% of individuals with Type 2 diabetes have impaired diastolic function (18), which is associated with reduced O_2 max in healthy subjects (32). Roy et al. (22) showed a lower resting and exercising in a combined group of Type 1 and 2 diabetic patients than in nondiabetic controls; however, it is not clear from their findings whether this difference was the result of diabetes or differences in fitness level.

Peripheral O₂ delivery and extraction may also be affected by Type 2 diabetes. Type 2 diabetic patients have impaired nitric oxide-induced vascular function (16, 34), which can result in impaired muscle blood flow during exercise (15). Moreover, Type 2 diabetic patients have reduced oxidative enzyme activity (26), increased percentage of type IIb fibers, and decreased percentage of type I fibers (13). However, these characteristics are similar to those in obese nondiabetic controls (13, 29) and may reflect a lack of fitness, rather than a specific effect of diabetes.

To clarify whether and peripheral mechanisms are responsible for the decreased O_2 max in individuals with Type 2 diabetes, this study compared O₂, , and calculated arterial-mixed venous (a-) O₂ difference in Type 2 diabetic patients with age-, weight-, and fitness-matched controls at maximal and submaximal exercise intensities.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects and screening. Eleven individuals with Type 2 diabetes and 12 nondiabetic individuals, aged 40-60 yr, completed the study. Only individuals with body mass index between 25 and 40 kg/m² who had performed fewer than three 20-min sessions per week of moderate-intensity physical activity for 2 yr were recruited for the study. After providing informed, written consent, all subjects underwent a medical screening protocol. Screening included a medical examination, resting transthoracic echocardiogram, 12-lead exercise electrocardiogram test (Bruce protocol), blood analyses of fasting glucose, glycosylated hemoglobin, serum lipids, and creatinine, and spot urine analysis for albumin-to-creatinine ratio. Subjects were excluded if they showed clinical or symptomatic evidence of cardiovascular or pulmonary disease, impaired systolic function, or valvular abnormalities on echocardiogram, had uncontrolled hypertension (i.e., had resting blood pressure >160/90 mmHg or were taking >1 antihypertensive medication), showed evidence of impaired renal function (urinary albumin-to-creatinine ratio > 2.5), or were taking insulin. Eight diabetic patients were taking diabetic medications (metformin-glipizide), and two diabetic patients were taking angiotensin-converting enzyme inhibitors for hypertension. No control subject was taking prescribed medication. For the diabetic group, the average duration of diagnosis of diabetes was 5.4 ± 3.1 yr. Subject characteristics are summarized in Table 1.

Experimental measurements and study design. At 1-2 wk after medical screening, subjects performed an incremental workload O_2 max test on a cycle ergometer (model ERG 900, Schiller) that maintains constant power output at different pedaling velocities. All testing was performed in the same laboratory at 22-23°C. Initial workloads were set at 25 or 50 W, depending on subject ability. Workload was increased by 25 W each 1-min stage, unless the subject did not believe he or she could perform a full-stage increase, in which case workload increased by 15 W. With use of this individualized protocol, all tests lasted 6-12 min. Breath-by-breath data were collected and analyzed on O₂ and CO₂ analyzers (models S-3A and CD-3A, respectively, Ametek), which were calibrated with room air and standardized gas containing 6% CO₂ before each test. It was assumed that O_2 max had been achieved when two of the following three criteria had been met: 1) <2 ml · kg¹ · min¹ increase in O₂ with increase in workload, 2) respiratory exchange ratio (RER) > 1.10, and 3) achievement of age-predicted heart rate (31). Failure to meet two of these criteria resulted in a retest or exclusion from further analysis.

Maximal (CO₂ rebreathing). After ~1 wk, maximal (_max) and submaximal were estimated noninvasively during cycling exercise by the CO₂ rebreathing equilibration method using the Fick equation, as originally performed by Collier (3) and described by Jones (9). Gas analyzers were calibrated with room air and standardized gas consisting of 13% CO₂ before each test. was measured during two 5-min bouts of exercise at 60 and 70% O_2 max as established during the O_2 max test. To achieve steady state, subjects cycled at a predetermined constant workload equaling 60% or 70% O_2 max until a plateau in O₂ occurred and heart rate (as measured by a Polar monitor) fluctuated by less than five beats in successive minutes. Once this happened, the subject briefly held his or her breath while inspired gas was switched to a 3-liter anesthetic bag that contained 12% CO₂-88% O₂ at 60% O_2 max and 13% CO₂-87% O₂ at 70% O_2 max, as recommended by Jones. End-tidal PCO₂ was measured and displayed using Chart version 4.0 software on a Maclab computer. Rebreathing continued until PCO₂ had reached a plateau or for 15 s. Stroke volume was calculated as /heart rate, and a- O₂ difference was calculated as O₂/. Because stroke volume is thought to plateau at ~40% O_2 max in nonelite athletes (1, 6), maximum stroke volume was estimated as the average of the two submaximal values. _max was determined by multiplying the maximal stroke volume by the heart rate obtained at O_2 max. With use of this method, a coefficient of variation (SD/mean × 100) of 4.5% has been established for three repeated estimates of _max in nondiabetic subjects in this laboratory.

Statistical comparisons. Unpaired Student's t-tests (2-tailed) were used to compare means between Type 2 diabetic and control groups. Univariate linear regression was performed to identify the relationship between continuous variables. P < 0.05 was considered statistically significant. Values are means ± SD.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Age, body composition, and blood and urinary markers were similar between groups with the exception of high-density lipoprotein cholesterol, which was lower in Type 2 diabetic patients, and glycosylated Hb, fasting glucose, and triglycerides, which were higher in Type 2 diabetic patients (P < 0.05). Seven subjects were excluded by medical screening or failure to achieve O_2 max. The resulting groups were 36% (diabetic group) and 58% (control group) female. The data are summarized in Table 1.

Maximal and relative hemodynamic results. Group comparisons at O_2 max and 60 and 70% O_2 max are summarized in Table 2. The control group achieved higher workloads at 60, 70, and 100% O_2 max than the diabetic group. Peak and submaximal O₂ were higher in the control group. RER was similar between groups at each intensity. Heart rate was similar between groups at peak exercise but was significantly lower in the diabetic group at 60 and 70% O_2 max. and stroke volume were similar between groups at each relative intensity. a- O₂ difference was similar between groups at 60% O_2 max (P = 0.12), tended to be higher in the control group at 70% O_2 max (P = 0.07), and was significantly higher in the control group at 100% O_2 max (P < 0.05).

Hemodynamic results at identical absolute submaximal workloads. The number of subjects in the control and diabetic groups completing incremental exercise stages at 50, 75, 100, 125, and 150 W was 12 and 11, 12 and 11, 12 and 9, 10 and 5, and 8 and 3, respectively. Figure 1A shows that O₂ (l/min) was higher in the control group at 75 and 125 W (1.03 ± 0.20 vs. 0.80 ± 0.20 and 1.52 ± 0.30 vs. 1.20 ± 0.20, P < 0.05) and tended to be higher at 50 W (0.80 ± 0.20 vs. 0.63 ± 0.20, P = 0.064), 100 W (1.24 ± 0.30 vs. 1.03 ± 0.20, P = 0.058), and 150 W (1.73 ± 0.40 vs. 1.42 ± 0.00, P = 0.061). RER was not significantly different (P > 0.05) at any workload (Fig. 1B). Heart rate was similar at lower workloads but was higher in the control group at 125 and 150 W (Fig. 1C).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Submaximal workload vs. O2 consumption (A), respiratory exchange ratio (B), and heart rate (C) in control and Type 2 diabetic patients during an incremental bicycle ergometer test. Values are means ± SD. * Significantly different from Type 2 diabetes group (P < 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Linear relationship between individual values for maximal O2 consumption vs. maximal cardiac output (A) and maximal arterial mixed venous (a-) O2 difference (B) in Type 2 diabetic patients.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

is generally thought to limit O_2 max because of the gross potential imbalance between skeletal muscle capacitance and _max (24). Thus, on the basis of O_2 max values, it is not surprising that both groups of subjects had low _max. However, the findings of this study provide indirect evidence that O_2 max is further reduced in sedentary Type 2 diabetic patients by an impairment in peripheral O₂ extraction. O_2 max was 30% lower, _max was similar, and maximal a- O₂ difference was 19% lower in diabetic patients than in control subjects matched for physical activity, body composition, and age. Furthermore, O_2 max was correlated with a- O₂ difference, but not with _max in Type 2 diabetic patients, suggesting that impaired maximal total body O₂ extraction contributed to lower O_2 max in Type 2 diabetic patients.

Some methodological factors should be considered when these results are interpreted. The data shown above are dependent on accurate determination of O_2 max. To the best of our ability, these data represented a "true" maximal value based on the fact that only subjects who achieved a plateau in O₂ and RER > 1.10 were included. Nonetheless, only two subjects achieved their age-predicted maximal heart rate, and this may have contributed to the low O_2 max in both groups. Because maximal heart rate is used to calculate _max, it is conceivable that _max values reported here are underestimated; however, the _max values obtained at 126 W (in diabetic patients) and 173 W (in controls) are consistent with previously reported submaximal measured by the CO₂ rebreathing method during steady-state exercise at 133 and 200 W in trained nondiabetic subjects (30), suggesting that the values reported here are typical of these exercise intensities.

Medications taken by the diabetic group may have affected peripheral blood flow during exercise. Five patients took "low doses" of sulfonylurea drugs, which may reduce peripheral vascular function (2), and four patients took a biguanidine drug, which may improve endothelial function (14). However, it is impossible to estimate the effect, if any, of these two medications in the diabetic group, particularly during exercise.

Although the results of the present study are similar to results of previous studies that have reported O_2 max to be 20-30% lower in Type 2 diabetic patients than in age-, fitness-, and weight-matched nondiabetic controls (7, 8, 10, 19, 20, 27), the O_2 max values reported here are lower than those reported previously, with one exception (19), in which only female subjects were studied. Although 58% of our control subjects were women, we believe that our control subjects may represent a less fit sample than has been previously reported. This contention is supported by the fact that _max, which is lower in less fit individuals (1), was comparable to that in a group of Type 1 and Type 2 diabetic patients (22) but significantly lower than in healthy, nontrained subjects with _max of 17-22 l/min (1, 22, 23).

Maximal heart rate was lower than age-predicted norms (31), which likely contributed to low _max in the present study; however, an attenuated increase in exercise stroke volume may also have contributed. Maximum stroke volume in both groups was ~70 ml/beat, which is considerably lower than previously reported values in sedentary controls (8, 23). In healthy young control subjects examined in this laboratory, average maximal stroke volume was 112 ml (unpublished findings). Had the control subjects in this study achieved similar stroke volumes, their _max would be comparable with those reported in previous studies in healthy controls. Instead, their stroke volumes were comparable to those reported after deconditioning by prolonged bed rest (8, 25), reaffirming our contention that these subjects represented a less aerobically fit population than previous investigations. We can only assume that resting values were normal on the basis of normal resting left ventricular end-diastolic and end-systolic dimensions and fractional shortening, which were similar in both groups.

The reasons for the difference in a- O₂ difference between groups can only be speculated from these data. The a- O₂ difference values in our ~50-yr-old diabetic group were only slightly higher than those reported in 60- to 69-yr-old men (28), suggesting that such low values may be normal for this age group; however, values in the age-matched control group were comparable to previous values in young subjects (1), suggesting that some characteristic of diabetes, independent of age and fitness, resulted in a lower a- O₂ difference.

Increases in mixed venous O₂ content, caused by a maldistribution of blood flow and/or reduced oxidative capacity of skeletal muscle, may explain this difference. Type 2 diabetic patients have impaired nitric oxide-mediated vascular function (34), which inhibits peripheral vasodilation (16). Maxwell and colleagues (15) showed that inhibition of endothelium-derived nitric oxide reduced the distribution of fluorescent microspheres to limb skeletal muscle after exercise. Furthermore, microvascular complications such as retinopathy and nephropathy, which are associated with impaired vascular function, have been associated with reduced exercise capacity in Type 2 diabetic patients (5). Thus the diabetic patients in this study may have had increased mixed venous O₂ content resulting from decreased blood flow to working muscles during cycling exercise. This contention is supported by recent findings that have shown that exercise training improves peak O₂ (12) and peripheral vascular function (11) in Type 2 diabetic patients.

It is difficult to understand why submaximal O₂ is lower in Type 2 diabetic patients. a- O₂ difference was reduced at 100% O_2 max and tended to be lower at 70% O_2 max (P = 0.07) in diabetic patients. According to the Fick equation, this suggests that diabetic patients should have required higher to meet the O₂ demand of submaximal activities. However, heart rate was lower in diabetic patients than in controls, and stroke volume at submaximal workloads was similar to that in controls. Although difficult to explain, these findings are similar to those of Regensteiner et al. (20), who reported that O₂ was lower during identical submaximal workloads in Type 2 diabetic patients. They showed that this resulted from slower O₂ and heart rate kinetic responses to increased submaximal workload (19).

Alterations in cellular properties of skeletal muscle may have affected submaximal O₂. Type 2 diabetic individuals have increased type IIb-to-type I fiber ratio (13) and lower oxidative enzyme activity (26, 29, 33) than nondiabetic subjects. These reductions in the oxidative capacity of skeletal muscle may have resulted in less O₂ being consumed by working muscles in diabetic patients during submaximal exercise. Theoretically, increases in RER would be expected because of the resultant increase in anaerobic metabolism to meet the energy demand of work. Although not significant, it is noteworthy that RER was higher in diabetic patients at every workload.

In summary, the results of this study confirm previous findings that sedentary individuals (diabetic and nondiabetic) have low O_2 max and _max. However, Type 2 diabetic patients also had lower maximal a- O₂ difference than nondiabetic individuals of similarly low fitness, which may have further reduced their O_2 max. The reasons for this are not known; however, impairment in peripheral vascular function and/or skeletal muscle O₂ extraction are areas for further investigation.