Thenar Muscle Oxygen Saturation Using Vascular Occlusion Test: A Novel Technique to Study Microcirculatory Abnormalities in Pediatric Heart Failure Patients

Abstract

Heart failure (HF) is associated with microcirculatory changes secondary to neuro-humoral imbalance, vascular stiffness and increased sympathetic tone. Near Infra-Red Spectroscopy (NIRS) derived Thenar muscle tissue oxygenation levels (StO₂) can provide an estimate of the functional status of microcirculation. There is a paucity of literature regarding evaluation of microcirculation in pediatric subjects with HF. We hypothesized that microcirculation and oxygen saturation dynamics as assessed by Thenar StO₂ levels using vascular occlusion test (VOT) would be altered in HF subjects and that these changes may correlate with the severity of heart failure. We prospectively enrolled 60 pediatric subjects (29 healthy control, 31 HF). Baseline StO₂ levels were measured using InSpectra™ StO₂ probe placed over the Thenar eminence of right hand, followed by a VOT for 3 min, during which the changes in StO₂ levels during the occlusion phase and post occlusion phase were recorded. Baseline Thenar StO₂ levels (72 ± 8 vs 76 ± 5, p = 0.02) and time to baseline StO₂ in seconds (150 ± 70 vs 200 ± 70, p = 0.007) were significantly lower in HF group compared to healthy control (HC). In addition, HF patients had a significantly lower trough StO₂ (37 ± 9 vs 42 ± 11%, p = 0.04) and peak StO₂ compared to HC (87 ± 8 vs 91 ± 5%, p = 0.01). However, there was no difference in the rate of desaturation, rate of resaturation or time to peak StO₂ levels in between the 2 groups. Significant correlation was present between baseline Thenar StO₂ levels and NYU Pediatric Heart Failure Index Score (NYU-PHFI) (p = 0.003). This study is the first to report an objective assessment of microcirculation and Thenar tissue oxygen dynamics in pediatric subjects with HF in comparison with HC. Our study suggests altered microcirculation and oxygenation patterns in these subjects as well as correlation with a validated pediatric heart failure clinical score. Large-scale prospective studies are needed to further study the utility of this novel technology in HF subjects.

Introduction

Pediatric heart failure is multifactorial secondary to congenital heart disease and cardiomyopathy or is due to acquired heart disease such as myocarditis, rheumatic fever, endocarditis, and Kawasaki disease. Hospitalizations related to heart failure impose a huge economic burden. In a recent analysis of the Healthcare Cost and Utilization Project Kids’ Inpatient Database and Nationwide Inpatient Sample of pediatric cardiomyopathy, heart failure-related hospitalizations were associated with increased length of stay, worsened outcomes, and an increase in cost greater than that associated with adult hospitalizations [1]. Clinical assessment of subjects in compensated heart failure may be challenging and subtle changes may not be evident. The status of heart failure in patients seen in an out of hospital setting is managed based on symptoms, heart failure clinical scales, clinical examination, brain natriuretic peptide (NT-BNP), and associated biomarkers of end-organ function. Heart failure evaluation by symptoms, examination, and radiologic studies has demonstrated poor sensitivity and specificity [2]. In addition, associated markers may not demonstrate remarkable changes in a patient’s clinical condition until the later stages of heart failure manifest, due to compensatory mechanisms. However, when the compensatory mechanisms are overwhelmed or fail in the setting of a new stressor (such as infection), there is a rapid deterioration in clinical status, limiting the chances for early optimization of medical therapy to improve outcomes.

Heart failure is associated with peripheral endothelial dysfunction and decreased peripheral vascular reactivity [3-5]. This dysfunction has been primarily attributed to vasoconstriction secondary to elevated intrinsic catecholamines, altered vasomotor tone, neuro-humoral changes, and increased vascular stiffness. The resulting vasoconstriction increases aortic impedance, and the failing ventricle is subjected to higher after-load [6]. Accumulating evidence suggests that microcirculatory changes play a significant role in the development and progression of heart failure. Inaccuracy in grading and failure to appropriately recognize heart failure severity with interval worsening is associated with higher mortality and increased frequency of re-hospitalization [7]. Currently recommended methods of invasive cardiac monitoring are not practical in all subjects with heart failure, as they present innate, unnecessary risk and are associated with higher costs [8]. There is an unmet need for improved, non-invasive evaluation, and understanding of heart failure pathophysiology, in order to optimize therapy, bolster clinical management, and improve HF outcomes.

Regional oxygen saturation monitoring (rSO₂) using Near Infrared Spectroscopy (NIRS) technology is now widely available and used in management of several pathophysiological states, including post-operative cardiac surgery subjects, sepsis, trauma, neuroprotection, and splanchnic ischemia [9]. NIRS technology non-invasively measures the attenuation of light by hemoglobin, utilizing a narrow spectrum of wavelengths which penetrate deep into tissue. This technology primarily utilizes wavelengths from 700 to 850 nm, maximizing the difference between oxy- and deoxy-hemoglobin and minimizing the influence of other chromophores. The Lambert Beer law is used to calculate hemoglobin and deoxy-hemoglobin absorption, and the difference between the absorption coefficients eventually allows for the calculation of tissue oxygenation levels.

Thenar muscle tissue oxygen saturation (StO₂) using NIRS technology has been studied in adult HF, sepsis, and trauma patients. It has been found to be a useful marker of regional oxygenation in adult CHF subjects with sepsis and has been shown to successfully predict mortality and outcomes among the sepsis cohort [10, 11]. Further, Thenar StO₂ levels have also been correlated to fluctuation in concordance with serum lactate levels in adult subjects with sepsis [12, 13]. In patients with chronic HF, adult investigations have demonstrated low baseline Thenar StO₂ levels improve after initiation of inotropic therapy [14].

To date, there have been no pediatric studies evaluating the utility of Thenar muscle StO₂ measurement in children with heart failure. We hypothesized that the baseline and peak Thenar StO₂ would be lower in subjects with HF and that the rate of desaturation (R_des), recorded during standard VOT, would be more pronounced in these patients, along with a delayed rate of reperfusion (R_res) in subjects with heart failure compared to healthy controls.

Methodology

Study Design

We undertook a single center, prospective, non-randomized, case control cohort study at UF Health Shands Children’s Hospital. Enrollment persisted over a period of 24 months from January 2015 to December 2016. The study was approved by the University of Florida IRB. During the study period, patients who presented for heart failure (HF) follow-up visits at our Congenital Heart Center were approached for enrollment. For our healthy control cohort, enrolled participants presented to the Pediatric Congenital Heart Center clinics for evaluation of heart murmur and/or chest pain, without any underlying cardiac disease on an echocardiogram. Children with acute deterioration in clinical status, active infection, vascular disease, autoimmune vasculitis, limb deformities, limb pain, bone disorders, and severe anemia were excluded.

Baseline Tissue Oxygen Saturation (StO₂) Measurement and VOT

The InSpectra™ StO₂ monitor, manufactured by Hutchinson Technology INC., was used to measure the baseline StO₂ level after applying the non-invasive probe to the skin on the Thenar eminence. After a stable baseline StO₂ (%) reading was obtained, the VOT was performed as described: a blood pressure cuff placed on the upper extremity was inflated to 40 mmHg above the systolic pressure and the rate of desaturation (R_des; % × s⁻¹) was recorded. After 3 min, the cuff pressure was released instantaneously and the rate of reperfusion (R_res; % × s⁻¹) was measured. The measurement continued until StO₂ returned to baseline. Figure 1 shows a representative graph of the measurement trend.

Fig. 1.

Typical StO₂ changes during the Vascular Occlusion Test (VOT)

Data Collection

Patient demographic data collected included age (months), gender, weight (kg), body mass index (kg/m²), primary diagnosis, and any secondary diagnoses. Clinical data was collected, including heart failure stage (NYU Pediatric Heart failure Index Score), most recent echocardiographic findings, and associated comorbidities. Results from the Thenar StO₂ monitor included baseline StO₂ (%), peak StO₂ (%), trough StO₂ (%), rate of desaturation R_des (%/s), rate of resaturation R_res (%/s), time to peak StO₂ (secs), and time to trough StO₂ (secs).

Statistics

A p value of < 0.05 was taken to convey significance, and all statistical analyses were performed using the R software package (The R Foundation for Statistical Computing, Vienna, V. 3.2.4). Changes in StO₂, SpO₂, and laboratory values were evaluated for normality, and distributions were explored, plotted, and tested for equality utilizing the Forsythe–Levene and Shapiro–Wilk tests. Parametric tests were run on normally distributed, continuous data, while non-parametric alternatives were employed to examine other relationships. χ² and Fisher’s exact tests were implemented for group comparison of categorical variables, while one-way ANOVA and t tests were implemented for the comparison of group means among continuous variables. A Bland–Altman plot was constructed to demonstrate agreement between variables which documented significant correlation, and either Person’s r or Spearman’s rho was employed for correlation analysis.

Results

We prospectively enrolled 60 subjects (29 HC, 31 HF) over a period of 24 months. Patient and control characteristics are summarized in Table 1. The mean age of the subjects was 10.3 ± 4.2 years in HC, and 12.7 ± 6.2 years in HF groups. The 2 groups were found to be similar with respect to age, body weight, body mass index (BMI), and body surface area (BSA). As expected, there was significant difference in the ejection fraction derived from Simpson method (EF, p = 0.0002), shortening fraction (SF, p = 0.0002), and heart rate (p = 0.0006) between the 2 groups.

Table 1.

Patient characteristics

Variable	Healthy control (n = 29, 35.4%)	Heart failure (n = 31, 37.8%)	p value
Male, n(%)	15 (52%)	15 (48%)	NS
Age (mean ± SD)	10.3 ± 4.2	12.7 ± 6.2	NS
Weight in kg (mean ± SD)	44.9 ± 25.4	55.7 ± 36.0	NS
BMI (mean ± SD)	19.6 ± 6.0	23.3 ± 9.1	NS
EF/% (mean ± SD)	63.9 ± 5.6	50.1 ± 16.3	.0002
SF (mean ± SD)	36.0 ± 4.4	26.5 ± 10.8	.0002
Heart rate (mean ± SD)	86.83 ± 17.32	100.1 ± 17.01	.0006
Systolic blood pressure (mean ± SD)	109 ± 14.34	122.8 ± 91.45	NS
Diastolic blood pressure (mean ± SD)	65.52 ± 11.73	68.23 ± 9.69	NS
Mean blood pressure (mean ± SD)	80.03 ± 11.85	86.43 ± 31.93	NS

Statistically significant values are given in bold (p < 0.05)

Baseline Thenar StO₂ levels were significantly lower in the HF group (72 ± 8 vs 76 ± 5, p = 0.02) as well as time to baseline StO₂ in seconds (150 ± 70 vs 200 ± 70, p = 0.007) (Table 2). In addition, HF patients had a significantly lower trough StO₂ (37 ± 9 vs 42 ± 11%, p = 0.04) and peak StO₂ compared to HC (87 ± 8 vs 91 ± 4%, p = 0.01). Despite this significant finding, there was no difference in the rate of desaturation, rate of resaturation, or time to peak StO₂ levels in between the 2 groups. Interestingly, a Bland–Altman plot illustrated moderate agreement between pulse oximetry derived systemic oxygen saturation (SpO2) and baseline Thenar muscle StO₂ levels (Fig. 2). Spearman’s rho revealed significant correlation between baseline Thenar StO₂ levels and NYU-PHFI scores (ρ = p = 0.003) (Fig. 3). Baseline StO₂ levels showed a significant negative correlation with serum aspartate aminotransferase (AST) levels (r = −0.624, p = 0.003) (Fig. 4). Similarly a significant negative correlation was found between rate of desaturation and serum hemoglobin levels as well as with tissue hemoglobin index (THI) (Fig. 4).

Table 2.

Thenar StO₂ in HC, HF groups

Variable	Healthy control (n = 29)	Heart failure (n = 31)	p value
Baseline StO2 (%)	76 ± 5	72 ± 8	0.02
Rdes (%/s)	0.19 ± 0.06	0.19 ± 0.05	NS
Trough StO2 (%)	42 ± 11	37 ± 9	0.04
Rres (%/s)	1.77 ± .71	1.87 ± .75	NS
Peak StO2 (%)	91 ± 4	87 ± 8	0.01
Time to peak StO2 (s)	31 ± 9	33 ± 19	NS
Time to baseline (sec)	200 ± 70	150 ± 70	0.007

Statistically significant values are given in bold (p < 0.05)

Fig. 2.

Bland-Altman Agreement SpO2 vs. StO2. A Bland–Altman plot demonstrating moderate agreement between traditional pulse oximetry and baseline Thenar StO₂ in pediatric patients with heart failure. (Bias = 21.61%; LOA = (μ± 1.96sd) 5.11%, 38.11%)

Fig. 3.

Correlation of Thenar StO2 and NYU-PHFI score. Spearman’s correlation reveals a significant association between NYU PHFI Score and baseline Thenar StO₂. When higher O₂ saturation was detected in peripheral tissue, enough evidence existed to support moderate correlation with a decrease in HF score. (n = 29, ρ = −0.526, p = 0.0034)

Fig. 4.

Correlations were examined between baseline Thenar tissue oxygenation levels and serum concentrations of aminotransferase enzymes (a ρ = − 0.624, p = 0.003), the rate of desaturation and serum hemoglobin concentration (b r = − 0.657, p = 0.002), and tissue hemoglobin index (c r = − 0.525, p = 0.003) in pediatric patients with heart failure

Discussion

This prospective case control cohort study evaluates the utility of measuring Thenar StO₂ levels in pediatric subjects with HF, in comparison with a group of HC. Our study, the first of its kind in this pediatric population, demonstrates the deranged microcirculation and oxygenation parameters in the pediatric HF group compared to HC.

Pediatric heart failure is associated with significant morbidity and mortality [1]. There is an imbalance between oxygen delivery (DO₂) and consumption (VO₂) in subjects with HF. In addition, significant microcirculatory disturbance is present in HF subjects due to reduced capillaries, arteriovenous shunting, and decreased flow. These microvascular abnormalities were found to be more severe in non-survivors [15]. It has been proposed that these changes in microcirculation are secondary to decreased oxidative muscle fibers, mitochondrial depletion, early anaerobic metabolism, imbalance between vasoconstrictor and vasodilator micro-vascular tone, and systemic neuro-humoral factors. Identification of micro-vascular abnormalities is clinically relevant, as previous studies have shown that these changes can improve with appropriate therapy over time [16]. Conventional tools like lactate, blood gas, pulse oximetry, and echocardiogram are useful tools; however, they measure blood oxygen levels or heart function and do not reflect abnormal micro-perfusion. Routine assessment of tissue oxygenation continues to be a challenge in clinical practice. An optimal tool is desperately needed to serially measure and quantify the microcirculatory changes and vascular dysfunction with HF and monitor response to therapy.

Near infrared spectroscopy (NIRS) derived is a non-invasive optical technique that can be used to measure tissue oxygenation status. Initially developed as a research tool, NIRS has recently become more popular for use in clinical practice, increasing the evidence base to support its incorporation [9]. Specifically, cerebral and renal rSO₂ monitoring is already common practice in pediatric cardiac ICUs and the same technology has been employed to measure regional muscle oxygenation status. Physiologically, the forearm (including Thenar musculature) is a predominant site of vasoconstriction in shock, making it a suitable site for rSO₂ measurement [17]. Using this knowledge, the measurement of microvascular oxygen utilization at the Thenar eminence may therefore detect subtle changes in vasoconstriction and endothelial dysfunction which may be otherwise undetectable by current global evaluation techniques [18]. Through the application of traditional rSO2 techniques in the Thenar musculature, dynamic assessment of oxygen consumption rates, endothelial function (reperfusion rate), and vascular reserve (reactive hyperemia) within skeletal muscle are now possible utilizing NIRS technology and a vascular occlusion test [19].

NIRS-derived cerebral and renal rSO₂ has been used and studied after pediatric cardiac surgery [20]. However, Thenar StO₂ levels have not been studied in children and there are no pediatric studies which evaluate the utility of Thenar muscle oxygenation status and VOT in subjects with HF. Thenar StO₂ has several advantages over cerebral and renal oxygen saturation monitoring. Thenar StO₂ monitoring device is portable with an easy-to-use clip on device which can be used in outpatient practice. In contrast, cerebral and renal rSO₂ monitoring devices are bulky, used in monitoring inpatient population and not available to monitor patients in out of hospital setting during follow-up. Dynamic monitoring of tissue oxygenation is possible with Thenar StO₂ monitor using the vascular occlusion test, which give additional data on local oxygen consumption (VO₂) and post-ischemic reperfusion and hyperemia. These variables assess and identify the state of the microcirculation and its response to vascular occlusion. These are particularly relevant in the long-term follow-up of these patients as heart failure is associated with microcirculatory abnormalities. This dynamic assessment of the microcirculation is not possible with cerebral or renal rSO₂ monitoring. Thenar StO₂ monitoring of heart failure patients in out of hospital setting could potentially identify worsening heart failure early and could aid in the management of these patients.

Regional oxygen saturation, as measured by Thenar StO₂, has been extensively studied in adult populations and is found to be altered in several disease states (heart failure, sepsis, trauma). Thenar StO₂ levels have been shown to be useful in predicting organ dysfunction and in prognostication of adult subjects with sepsis [10, 11], as well as in discriminating adult trauma patients with or without shock, and predicting mortality and organ dysfunction after trauma [13, 21, 22]. This novel technology has been evaluated in adult heart failure patients in multiple studies. Further, Thenar StO₂ levels have been validated against invasive measurements following cardiac surgery [23]. In subjects with left ventricular failure without sepsis, resting StO₂ levels showed good correlation with mixed venous oxygen saturation levels [24]. Thenar StO₂ has previously shown to be low in adult heart failure subjects and has been associated with later development of adverse outcomes [7]. To the best of our knowledge, there are no pediatric studies evaluating the use of Thenar StO₂, particularly in the heart failure population.

We evaluated this novel, non-invasive, quantitative technology for the assessment of microcirculation in pediatric subjects with heart failure. NIRS-derived StO₂ predominantly reflects venous blood oxygen saturation and thus is a marker of tissue oxygen delivery-consumption balance. In addition to Thenar muscle oxygenation levels, we evaluated the vascular reserve using the VOT, as well as oxygen dynamics in peripheral microcirculation using various components of the VOT curve (desaturation and reperfusion slopes). The application of intra-measurement occlusion techniques eliminates the influence of blood flow rate on the Thenar StO₂ values, ensuring the changes are reflective of microcirculation and associated abnormalities. Our study results demonstrate that tissue oxygenation is significantly impaired in HF subjects, resulting in lower baseline, peak, and trough StO₂, similar to previous adult studies [25, 26]. Our study was not designed to correlate Thenar StO₂ levels with ScvO2 levels as the study was conducted on patients in an outpatient clinic setting. We did however study the relation between systemic oxygen saturation measured by pulse oximetry and Thenar StO₂ levels. A Bland–Altman plot constructed from study subject measurements demonstrated moderate agreement between Thenar StO₂ levels and systemic oxygen saturation.

Although there is currently no single test available to diagnose heart failure severity in children, the NYU-PHFI score has recently become a useful clinical estimation tool that has been validated in pediatric populations. The NYU-PHFI is a weighted, linear combination of scores based on physiologic indicators and medical regimen. It carries a range of possible scores from zero (no heart failure) to 30 (severe heart failure) [27]. The Median NYU-PHFI score in our study group was 10 (range 3–23). After descriptive investigations and normality testing, either Pearson’s r or Spearman’s rho was employed to evaluate any existing correlation between StO₂ and our dependent variables. Unsurprisingly, a moderate, yet significant linear relationship was found between NYU-PHFI score and baseline Thenar StO₂ levels (ρ = − 0.497,p = 0.006), indicating that baseline Thenar StO₂% increases as the clinically defined staging of HF decreases in severity (Fig. 3). When baseline StO₂ was correlated with serum aminotransferase (AST) levels, additional significance was found among coefficients, indicating StO₂% tends to remain high as concentrations of AST fall in serum (ρ = −0.624,p = 0.003). Additionally, the rate of desaturation was also moderately, yet significantly correlated with the concentration of serum hemoglobin obtained at the time of the StO₂ measurement (r = − 0.657, p = 0.002, n = 19), as well as the Tissue Hemoglobin Index (THI), obtained via the Inspectra monitor™ (r = −0.525, p = 0.003). Taken together, these results provide enough evidence to suggest that Thenar oxygenation levels display significant correlation with the validated NYU-PHFI score, available hemoglobin AST, and THI, demonstrating that StO₂ measurements may confer some benefit when used in conjunction with other available parameters to identify and manage pediatric heart failure. These findings bolster a growing body of support for the impact of Thenar StO₂′s non-invasive nature, bedside availability, ability to evaluate vascular reserve, and various components of microcirculation; continuing to make it an attractive option as an adjunct to currently available tools.

In our study, pediatric heart failure subjects had a shorter time to baseline saturation compared to healthy control subjects. After the release of pneumatic pressure in vascular occlusion test, there is reactive hyperemia where the StO₂ increases over baseline levels, indicating post-ischemic vasodilatation and capillary recruitment. In heart failure patients, this duration is shorter probably because there is inadequate post-ischemic vasodilatation and capillary recruitment. The likely explanation for this is multifactorial involving endothelial dysfunction, neurohumoral factors, and elevated systemic vascular resistance due to chronic heart failure. This can lead to inadequate vascular reserve which causes a shorter reperfusion phase. In healthy subjects, the vascular reserve is much higher and consequently there is a longer time needed to return to baseline as compared to heart failure subjects. Similar findings with lower reperfusion indicating low vascular reserve in heart failure have been demonstrated previously in adult patients [14].

Limitations

Thenar StO₂ levels can be affected by skin pigmentation or by myoglobin interference. Our study contained a small number of subjects and was observational in nature, making it difficult to draw strong conclusions. Our study lacked longitudinal follow-up and was not designed to look at the adverse events or re-hospitalizations. Further research is required to refine and further elucidate the innate numerical relationships between StO₂ and tissue micro perfusion.

Conclusions

Our study results suggest that Thenar muscle microcirculation and oxygen saturation dynamics, as assessed using NIRS derived tissue oxygen saturation and the vascular occlusion test, is abnormal in pediatric patients with heart failure. In addition, lower Thenar oxygenation levels correlate with heart failure severity. Future large-scale studies are needed to evaluate the utility of this technology in recognizing worsening heart failure, in prognostication, and its role in improving outcomes.