Olson, Thomas P.

Introduction Heart failure (HF) patients demonstrate augmented ventilation for a given metabolic demand which results in increased respiratory muscle work and subsequent competition for blood flow between the respiratory and locomotor muscle beds. Unloading the respiratory muscles during submaximal exercise has been shown to increase leg oxygen uptake (VO 2 ) in HF patients due increased leg blood flow via increased cardiac output and redistribution of cardiac output from the respiratory to locomotor muscles. It has been suggested that increased leg VO 2 following respiratory muscle unloading in HF during submaximal exercise is due to augmented convective O 2 transport in HF; however, it is unclear if diffusive O 2 transport also increases with respiratory muscle unloading. We hypothesized that respiratory muscle unloading during submaximal exercise will lead to increased leg VO 2 due to combined increases in convective and diffusive O 2 transport. Methods Ten HF patients with reduced systolic function (6M/4W; 53±15 yrs; BMI: 29±6 kg/m 2 ) were recruited. Patients performed two submaximal exercise bouts (i.e. with and without respiratory muscle unloading (via mechanical ventilator) (RM unloading and control (CTL), respectively)). During exercise, leg blood flow (Q L ) was measured via thermodilution. Intrathoracic pressure was measured via esophageal balloon. Radial arterial and femoral venous blood gases were measured. Blood gases were used to calculate arterial and venous oxygen content (CaO 2 and CvO 2 , respectively). Leg VO 2 was calculated as Q L multiplied by the difference between CaO 2 ‐CvO 2 . Leg O 2 delivery was calculated as Q L multiplied by CaO 2 . Leg diffusion capacity (DO 2 ) was calculated as leg VO 2 divided by femoral venous partial pressure of O 2 . Results From CTL to RM unloading, leg VO 2 increased during submaximal exercise (0.7±0.3 vs. 1.0±0.4 L/min, p<0.01, respectively). From CTL to RM unloading, leg O 2 delivery increased during submaximal exercise (0.9±0.4 vs. 1.4±0.5 L/min, p<0.01, respectively), whereas CaO 2 ‐CvO 2 was not different (14.1±3.0 vs. 14.2±3.2 mL/dL, p=0.57, respectively). From CTL to RM unloading, leg DO 2 increased during submaximal exercise (30.4±10.8 vs. 49.7±18.6 mL/mmHg, p<0.01, respectively). There was a negative relationship between the degree of RM unloading (i.e. percent decrease in mean inspiratory esophageal pressure) and percent increase in leg DO 2 with RM unloading (r= −0.77, p<0.01). Conclusion Our results demonstrate that unloading the respiratory muscles leads to increased leg VO 2 due to convective and diffusive O 2 transport during submaximal exercise in HF patients. These findings suggest that the naturally occurring respiratory muscle work in HF compromises convective and diffusive O 2 transport and thus constrains leg VO 2 during submaximal exercise in HF. Support or Funding Information NIH HL126638 AHA 18POST3990251 This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .

Background Hypertension (HTN) has a strong heritable component and the effectiveness of the treatment for HTN has been shown to be influenced by monogenic guidance. Each of the common HTN pharmacotherapies is ~50% effective in the population and layering blood pressure (BP) medications increases side effects and decreases medication adherence. HTN is a complex disease with primary interactions between the heart, vasculature, and kidney. Known and common genetic variants of each these organs have well‐described mechanisms. Despite the established bell‐curve response to treatment, the complexity of HTN, and functional variants of the major organ systems in HTN: no guidance, to date, utilizes a multi‐gene approach to guide BP therapy. Purpose To assess the association between a weighted multi‐gene algorithm and BP control in patients with HTN. Methods Non‐smokers with a family history of HTN were included in the analysis (n=384; age=61.0±0.9, 11% non‐white). 17 genotypes (ADRB1(2 SNPs), ADRB2 (2SNPs), CYP2D6, WNK (3SNPs), SLC12A3, SCNN1A, alpha‐adducin, renin, angiotensin (3 SNPs), ACE, and the angiotensin receptor) were weighted according to previous effect size in the literature (ΔBP when on the target therapy vs. not) as well as the number of peer‐reviewed papers of quality supporting each functional variant. After weighting, each genotype was entered into an algorithm that assessed pairing (e.g. homozygosity of functional ADRB1 is responsive to β‐blockade, heterozygosity is less responsive, and homozygosity of non‐functional is least responsive) within and between organ systems to predict the pharmacotherapy with greatest impact for each individual patient. Each pharmacotherapy was ranked from 1–4 as most to least likely to respond based on the algorithmic assessment of individual patient's genotypes. Three‐years of data was assessed at six‐month intervals for BP and medication history. Current and past pharmacotherapies were also assessed via chart review. Results There was no difference in initial BP at diagnosis between groups matching the top drug recommendation using the multi‐gene weighted algorithm (n=92) vs. those who did not match (n=292). However, from diagnosis to nadir, patients who matched the primary recommendation had a significantly greater drop in BP when compared to patients who did not (ΔSBP= −39.2±2.4 vs. −32.1±1.3mmHg , ΔDBP= −19.4±1.1 vs. −14.0±1.3mmHg , ΔMAP= −24.4±2.1 vs. −19.0±1.2mmHg, respectively, p<0.05 for SBP and DBP). Further, the difference between diagnosis to current 1yr. average BP was lower in the group that matched the top recommendation (ΔSBP= −33.2±2.3 vs. −27.4±1.2mmHg, ΔDBP=−14.8±1.1 vs. −11.5±1.2, ΔMAP= −21.2±2.3 vs. −15.6±1.8, respectively, p<0.05 for SBP and MAP). Conclusions These data suggest an association between a weighted multi‐gene algorithm on the BP response to pharmacotherapy. This magnitude of difference has previously been shown to be associated with a significant reduction in heart attack and stroke risk. Support or Funding Information Geneticure, Inc. This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .

The classic method for determining alveolar‐capillary membrane conductance (D M ) and pulmonary capillary blood volume (V C ) involves measurement of lung diffusing capacity for carbon monoxide (D LCO ) at multiple inspired O 2 tensions (F I O 2 ). Recently, the technique has been simplified by simultaneously measuring D LNO (nitric oxide) to determine values at a single F I O 2 . Controversy exists over equations for calculation of θ CO (reaction rate of CO with Hgb) and the correct diffusion ratio for NO and CO (α) in the calculation of D MCO . The purpose of this study was to investigate how various published equations and ratios influence D M & V C using both the multiple and single F I O 2 methods. Eight healthy males were studied. D LCO & D LNO were measured using the rebreathe method (inspired vol. ~75% of inspiratory capacity, F I O 2 18, 35, & 55%). D M & V C were first calculated using five reported equations for θ CO . Once D M & V C were optimized based on reported predicted values, the D LNO method was compared using a range of α values. Using the multiple F I O 2 method, depending on the equation for θ CO , D M ranged from 47 to 103 ml/min/mmHg, and V C from 74 to 102 ml. For the calculation of D M & V C using the D LNO method, α ranged from 1.5 to 4.0 in order to optimize D M & V C values to the classic method. The determination of D M & V C is critically dependent on the equations used for θ CO which in turn affects the optimal ratio used to correct the D LNO method. NIH HL71478

Pulmonary function (PF) may be influenced by a deterioration of cardiac function. As such, resting measures of PF may be prognostically useful in heart failure patients (HF). This study examined the prognostic value of resting PF in a cohort of HF. This included 113 patients (enrolled 1999‐2005, EF=28±11%; age=55±12yr; 62% male) followed for 64±34mo with death confirmed via the SSI & Mayo Clinic registry. PF included, forced vital capacity, FVC; diffusing capacity, DLCO; forced expiratory volume, FEV1; alveolar volume, VA. Patients were divided in tertiles according to PF with survival analysis via Log‐rank Mantel‐Cox Test with Chi squared analysis (Chi‐sq). Groups for FVC: 1)=80, 2)81‐97, 3)=98, %pre (Chi‐sq 21.5, p<0.01). Bonferonni correction (BC) suggested differences (diff) between groups 1 & 2 and 1 & 3 (p=0.01) with no diff (ND) between 2 & 3 (p=0.62). Groups for FEV1: 1)=77, 2)78‐95, 3)=96, %pre (Chi‐sq 16.3, p<0.01). BC suggested diff between groups 1 & 2 (p=0.02) and 1 & 3 (p<0.01). Groups for DLCO: 1)=75, 2)76‐90, 3)=91, %pre (Chi‐sq=8.2, p=0.02). BC suggested diff between groups 1 & 3 (p=0.003) with ND between 1 & 2 or 2 & 3 (p>0.05). Groups for VA: 1)=87, 2)88‐97, 3)=98, %pre (Chi‐sq=8.9, p=0.01). BC suggested diff between 1 & 3 (p<0.01) with ND between 1 & 2 or 2 & 3 (p>0.05). In a well defined cohort of HF patients, resting measures of PF appear to be predictive of all cause mortality. NIH HL71478

The ability to work efficiently at altitude is impaired by the reduced arterial oxygen tension. Acetazolamide may improve systemic saturation (SaO 2 ) by stimulating ventilation but increases the work of breathing. Efforts have focused on the potential role of sildenafil in improving exercise performance in high altitude environment. The purpose of this study was to determine the effects of acetazolamide and sildenafil on oxygen uptake (VO 2 ) and breathing efficiency during submaximal hypoxic exercise. Sixteen volunteers received sildenafil 40 mg (S), acetazolamide 125 mg (A) or matching placebo (P) tid × four doses in double‐blind fashion starting 8 hours prior to hypoxic exposure over 3 separate overnight stays in an altitude tent with an O 2 concentration of 12.5%, and then performed hypoxic submaximal exercise on a stationary bicycle for 10 min at a workload of 40% of VO 2 peak. SaO 2 decreased with exercise but there was no difference in exercising SaO 2 or VO 2 between conditions. As expected, acetazolamide increased ventilation compared to placebo and sildenafil. Breathing efficiency as measured by Ve/VCO 2 decreased with acetazolamide in comparison to placebo and sildenafil (P: 42.6±1.2, S: 41.9±1.6, A: 46.7±1.1, p<0.01). Sildenafil did not change breathing efficiency while acetazolamide reduced breathing efficiency during hypoxic exercise. This study was funded by Mayo Foundation and NIH Grant No. HL71478.

Reduced partial pressures of inspired O 2 may influence the bronchial circulation through a number of mechanisms (e.g. altered pulmonary and systemic vascular pressures, lung fluid regulation, airway cooling & drying from increased ventilation). The present study examined the impact of acute, graded hypoxia on bronchial blood flow (Q aw ) in eight healthy adults (28 ± 9 years). Subjects were exposed to inspired O 2 levels resulting in arterial O 2 saturation (SaO 2 ) values of 90% and 80% for 15 min each. To control for the mild hyperventilation that accompanies hypoxia, subjects then mimicked the 80% SaO 2 ventilation (V E ) rate with 21% FIO 2 and again with CO 2 added to the inspirate to maintain normocapnia. Q aw and cardiac output (CO) were measured using soluble gas techniques. Also measured were alveolar‐capillary conductance (D m , index of lung fluid balance), pulmonary capillary blood volume (V c ), and systemic mean (MAP) and pulmonary arterial pressures (PAP). Hypoxia resulted in minimal change in MAP, a small rise in PAP, minimal change in CO and V c , and a rise in D m . There was also a non significant decrease in Q aw with hypoxia. Mild hyperventilation resulted in a significant fall in Q aw , but when normocapnia was maintained, Q aw returned to baseline values. These data suggest that acute, graded hypoxic exposure does not significantly impact Q aw , however, the hypocapnia associated with mild hyperventilation reduces Q aw .

Heart failure (HF) patients have reduced cardiac reserve and increased work of breathing (WOB)., Increased demand for locomotor muscle blood flow (i.e. activities of daily living) may result in competition between respiratory and locomotor vascular beds. We hypothesized that HF patients would demonstrate improved locomotor blood flow with respiratory muscle unloading during moderate activity. We studied 10 HF patients (EF=30±8%) and 10 controls during two 10 min. cycling sessions at 60% VO 2 max. Session 1 (S1): 5 min of normal breathing (NB) and 5 min respiratory muscle unloading with a ventilator. Session 2 (S2): 5 min NB and 5 min of respiratory muscle loading with inspiratory resistance. Measurements included: leg blood flow (LBF), cardiac output (CO), and inspiratory pleural pressure time integral as an index of the WOB. During S1, WOB was reduced similarly in both groups (HF: 50±60%; CTL: 50±56%, p<0.05 for both). In HF, CO was reduced (9.6±1.4 vs 8.2±2.1 L/min, p<0.05) and LBF increased (4.9±2.6 vs 6.2±2.9 L/min, p<0.01). In CTL, there was no change in CO (14.7±3.1 vs 14.8±5.0 L/min) or LBF (10.9±5.6 vs 10.3±5.2 L/min). During S2, WOB was increased similarly in both groups (HF: 111±75%, CTL: 110±85%, p<0.01 for both). In HF, there was no change in CO (10.0±1.3 vs 10.3±2.5 L/min) or LBF (5.0±1.8 vs 4.9±1.4 L/min). In CTL, CO increased (15.4±4.3 vs 16.9±4.7 L/min, p<0.01) with no change in LBF (10.7±5.3 vs 10.3±5.6 L/min). These data suggest that the respiratory muscles of HF patients preferentially recruit blood flow away from the locomotor muscles during moderate intensity, whole body exercise.

Pulmonary hypertension is common in patients with heart failure (HF); however, for a given degree of left ventricular dysfunction, the range in pulmonary artery pressures (PAP) is large. Polymorphisms of a number of candidate genes have been implicated in contributing to smooth muscle dysfunction and remodeling of the pulmonary vasculature, including the serotonin transporter (5HTT) gene. Patients with HF may have mild tissue hypoxia, which may alter expression of 5HTT. This study examined the influence of a repeat length polymorphism in the promoter region of the 5HTT gene on differences in PAP in HF patients. Ninety HF patients (age=55±14yrs; LVEF=28±10%; NYHA=2.0±0.9) and 90 age and gender matched controls (CTL, age=51±15yrs; LVEF=63±7%) were recruited. Heart failure and CTL subjects were divided into 3 genotype groups, SS‐homozygotes for the short variant, LL‐homozygotes for the long variant and LS‐heterozygous condition. For both HF and CTL, SS n=20; LS n=41; LL n=29). PAP was calculated from the tricuspid regurgitant velocity. Age and gender did not differ between HF and CTL or among genotype groups. PAP was higher in HF compared to CTL (38.9±12.5 vs 27.7±8.2 mmHg, p<0.01, respectively) with a significant genotype effect in HF (SS=34.2±8.6 vs LS=38.2±11.4 vs LL=43.2±15.2 mmHg, p<0.05), but not in CTL. The HF group also demonstrated elevated left atrial diameter compared to CTL (HF 48.0±8.7 vs 32.8±4.8 mm, p<0.01, respectively) although no difference was seen within the HF group across genotypes. These results suggest that the LL variant of 5HTT is associated with elevated PAP in HF but not CTL. HL71478 and HL07111.

BACKGROUND Increased activation of Group III/IV skeletal muscle afferents is suggested to play a role in exercise intolerance in heart failure (HF). PURPOSE The purpose of this study was to examine if skeletal muscle afferent feedback inhibition improves exercise tolerance in HF. METHODS 8 HF patients [59±8 yr, mean±SD] participated in two randomized incremental recumbent cycling exercise sessions to peak capacity (peak oxygen consumption, VO 2 ). Session 1: The synthetic opioid fentanyl (50 mcg), known to block afferent feedback, was injected intrathecally (INJ) in the lumbar spine (L3‐L4 intervertebral space) prior to exercise. Session 2: Participants were prepared similarly with a sham injection (placebo, PLA). Gas exchange, dyspnea and RPE were measured. RESULTS HF patients exercised for a longer duration (Δ3±3 min, p=0.02) and reached a higher power output (PLA:109±30 vs. INJ:128±34 W, p=0.02) with locomotor afferent inhibition. Additionally, peak VO 2 was non‐significantly greater (PLA:18.7±3.3 vs. INJ:21.8±4.4 ml/kg/min, p=0.10). With INJ workload matched to peak PLA workload, RPE was reduced (PLA:18±2 vs. INJ:15±2, respectively, p=0.005) and dyspnea was lower (PLA:7.5±2 vs. INJ:4.2±1, p=0.003) with no difference in VO 2 (p>0.05). CONCLUSION Inhibition of group III/IV locomotor muscle afferent feedback contributes to improved exercise tolerance and exertional symptoms in HF patients.

Background Locomotor muscle group III/IV afferent feedback contributes to sympathoexcitation. Sub‐systolic lower limb occlusion may stimulate this reflex. Purpose To examine the cardiovascular response to lower limb venous occlusion in healthy humans during rest and exercise. Methods Nineteen adults (9 men, 25±5 yr; mean±SD) participated in two visits. V1: 3 min seated rest followed by randomized bilateral thigh tourniquet inflation to 20, 40, 60, 80, 100 mmHg for 2 min with 2 min deflation between occlusions. V2: Recumbent cycle ergometer exercise at 30% peak work with occlusion described in V1. We measured mean, systolic and diastolic pressure, cardiac output, heart rate, and stroke volume (MAP, SBP, DBP, CO, HR, and SV, respectively). Results At rest, MAP (85±18 mmHg) increased at 100 mmHg (Δ4±5, p<0.05) mediated by increased DBP (71±15 mmHg) at 100 mmHg (Δ5±4, p<0.05) with no change in SBP. During exercise, MAP (89±11 mmHg) increased at 60 (Δ7±7), 80 (Δ14±8), and 100 (Δ16±8) mmHg occlusions via increased SBP (133±22 mmHg) at 60 (Δ12±14), 80 (Δ20±13), and 100 (Δ26±13) mmHg and DBP (68±9 mmHg) at 60 (Δ5±5), 80 (Δ10±7), and 100 (Δ12±7) (p<0.05, all). Although, CO and SV did not change, HR (108±11 bpm) increased at 40 (Δ5±4), 60 (Δ8±5), 80 (Δ13±7), and 100 mmHg (Δ15±7) (p<0.05, all). Conclusion Stimulation of locomotor muscle group III/IV afferents via sub‐systolic occlusion contributes to cardiovascular control.