Flow ratios to express results obtained with the human in vivo ‘perfused forearm technique’

Abstract

Aims

To determine the accuracy of forearm blood flow (FBF) ratio (flow in infused arm/flow in control arm) to detect unilateral increases in forearm blood flow.

Methods

In nine healthy male volunteers, we measured the effect of infusion of saline into the brachial artery at a rate of 2 ml/100 ml forearm min⁻¹ on FBF ratio during control, mental arithmetic (MAR) and lower body negative pressure (LBNP) at −40 mmHg.

Results

Saline infusion increased FBF ratio from baseline by 115.9±17.4, 82.0±19.0 and 159.6±53.3% for control, MAR and LBNP, respectively (P<0.05 for MAR vs control).

Conclusions

FBF ratio may underestimate unilateral increases in forearm blood flow during simultaneous mental arousal.

Introduction

Recently, Petrie et al. advocated the use of the percentage change in forearm blood flow (FBF)-ratio (flow in infused arm/flow in noninfused arm) from baseline to express results obtained when vasoactive substances are infused into the brachial artery at subsystemic doses [1]. They argued that the use of FBF ratios will increase reproducibility and therefore allows the use of a smaller sample size as compared with alternative computations like responses in FBF or forearm vascular resistance obtained from both arms separately. However, apart from reproducibility, the precision of an observation is also determined by possible systematic error (bias). This study was conducted to assess potential bias in FBF ratios to express local changes in flow.

Methods

Subjects

This study was performed in nine healthy male nonsmoking normotensive volunteers, aged 31.2±11.7 (s.d.) years, with a mean forearm volume of 1033±160.2 (s.d.) ml. Informed consent was obtained from all volunteers. The study was approved by the local ethics committee.

Protocol

All volunteers came into the laboratory in the afternoon after a 24 h abstinence from caffeine or alcohol containing beverages and after a 10 h abstinence from any food intake. The brachial artery of the nondominant arm was cannulated with a 20 gauge catheter (Angiocath, Deseret Medical, Becton Dickinson, Sandy, Utah, USA) for infusion of saline (NaCl 0.9%).

Thirty minutes after cannulation, FBF was measured at both arms simultaneously using strain gauge plethysmography (Hokanson E-20, D.E. Hokanson, Washington D.C., USA). All forearm blood flows were measured three times min⁻¹ while wrist cuffs were inflated at 200 mmHg to exclude the hand blood flow [2]. Recordings were performed at baseline (4 min), during infusion of NaCl 0.9% into the brachial artery at a rate of 2 ml/100 ml forearm min⁻¹ (8 min) and during recovery (4 min). The saline infusion was repeated twice at 25 min intervals to allow forearm blood flow to return to baseline. During the last 4 min of two of these three saline infusions, a bilateral change in forearm flow was induced (‘disturbance’) by performing either a mental arithmetic (MAR, systemic disturbance resulting in forearm vasodilation) or lower body negative pressure of −40 mmHg (LBNP; systemic disturbance resulting in forearm vasoconstriction). The order of control infusion and disturbances was randomized.

The rapid infusion of saline was accomplished using two infusion pumps (IVAC 560, IVAC corporation, San Diego, California, USA). The performance of these pumps was checked after completion of the study in each volunteer. After the intra-arterial catheter had been removed at the end of the study, both pumps were reconnected to the catheter. To simulate intra-arterial blood pressure, which comprises considerable afterload to the infusion pumps, an extra resistance was built in between pumps and catheter, resulting in intraluminal pressures of 151 and 163 mmHg at flows of 20 and 30 ml min⁻¹, respectively. To measure actual infusion rate as achieved by the pumps during this artificial condition, saline was collected on a balance and weighed cummulatively at 30 s intervals. The pumps produced a reliable and steady flow under this artificial condition: after 8 min 160.8±0.04 (s.d.) and 243±0.2 ml had been infused at pump rates of 20 and 30 ml min⁻¹, respectively.

Statistics

In each volunteer, forearm blood flow ratios were calculated for each pair of simultaneous blood flow measurements (FBF in infused arm/FBF in noninfused arm) and averaged for each 4 min baseline period (12 flow ratios) and six subsequent 2 min intervals (six flow ratios per interval). Responses were calculated as the percentage change from baseline for each subsequent two minute interval. Group averaged results are presented as mean±s.e. mean. anova for repeated measurements was used to calculate levels of significance for differences in responses between the two disturbances (MAR and LBNP) with control. A two-sided P value <0.05 was considered significant.

Results

After 4 min, infusion of intra-arterial saline at a rate of 2 ml/100 ml forearm min⁻¹ increased forearm blood flow in the infused forearm from 3.2±0.7 to 6.9±0.7 ml/100 ml forearm min⁻¹ for control, from 2.7±0.5 to 6.5±0.7 ml/100 ml forearm min⁻¹ prior to MAR and from 2.8±0.5 to and 6.8±0.8 ml/100 ml forearm min⁻¹ prior to LBNP (P = NS for MAR or LBNP vs control). The intra-arterial saline infusion did not affect blood flow in the contralateral arm (see Figure 1).

Figure 1.

Course in forearm blood flow expressed as percentage response from baseline. *: significant difference between mental arithmetic and control; #: significant difference between LBNP and control (P<0.05; anova for repeated measurements with disturbance (control (•), mental arithmetic (▪), lower body negative pressure (▴)) as within-subject factor).

At the end of the saline infusions, forearm blood flow was 6.7±0.9, 8.1±0.8 and 3.8±0.6 ml/100 ml forearm min⁻¹ during control, MAR and LBNP, respectively (P<0.05 in response from baseline for MAR or LBNP vs control). In the noninfused arm, MAR increased FBF from 3.4±0.7 to 5.4±0.8 ml/100 ml forearm min⁻¹ (P<0.05) and LBNP reduced FBF from 3.2±0.7 to 2.1±0.3 ml/100 ml forearm min⁻¹ (see Figure 1).

After 4 min of saline infusion, FBF ratio increased from 1.2±0.2 to 2.6±0.4 during control, from 1.0±0.1 to 2.4±0.4 prior to MAR, and from 1.0±0.2 to 2.7±0.5 prior to LBNP (P = NS for MAR or LBNP vs control). At the end of the 8 min saline infusion, the FBF ratio was 2.4±0.4, 1.8±0.3 and 2.4±0.5 for control, MAR and LBNP, respectively (P<0.05 for MAR vs control). Similar results were obtained when percentage responses in FBF ratio were calculated (see Figure 1). After cessation of saline infusion and systemic disturbance, the forearm blood flow ratio quickly returned to baseline (see Figure 2).

Figure 2.

Course in forearm blood flow ratio expressed as percentage response from baseline. *: significant difference between mental arithmetic and control (P<0.05; anova for repeated measurements with disturbance (control (•), mental arithmetic (˘s), lower body negative pressure (▴)) as within-subject factor).

Discussion

The use of flow ratios has been advocated to correct for bilateral fluctuations in forearm blood flow that occur randomly in response to stimuli that are not directly related to the interventions in the infused forearm, for example mental arousal. It is assumed that systemic perturbations of blood flow have similar effects on FBF in infused-and contralateral forearm. Furthermore, the systemic perturbation in FBF should not interfere with the local intervention in the infused forearm. To validate these assumptions, we increased forearm flow with a fixed amount and calculated forearm blood flow ratios during control and during perturbations of bilateral forearm blood flow. For this purpose, we used a saline infusion instead of a pharmacological intervention as the effect of vasodilator infusions into the brachial artery depends on FBF since final plasma concentrations of the vasodilator substance are influenced by FBF.

Forearm flow increased more than expected from the amount of saline that was infused into the brachial artery, indicating that the saline infusion slightly reduced forearm vascular tone. Possible explanations for this observation include increased shear stress or reduced oxygen delivery to the infused forearm. Both hypoxia and shear stress are known to reduce vascular tone [3–5].

The main finding of this study is that the increase in flow ratio in response to saline infusion was reduced during simultaneous MAR. This observation indicates a different effect of MAR on vascular tone in infused and contralateral arm and suggests that a local increase in blood flow is underestimated by simultaneous mental arousal when results are expressed as relative changes in FBF ratio. Of course, the used systemic disturbances were substantial in our study. Bias in FBF ratio may be smaller when forearm studies are appropriately performed in a quiet room, reducing mental arousal as much as possible.

Several mechanisms may account for the observed effect of MAR on FBF ratio. The intra-arterial saline increased FBF which could have reduced MAR-induced vasodilation nonspecifically. Furthermore, a specific interaction between the intra-arterial saline infusion and mental arousal may have occurred. As mentioned above, shear stress and reduced oxygen supply to the infused forearm may have contributed to the increased forearm blood flow during intra-arterial saline infusion. Both hypoxia and shear stress induce endothelial nitric oxide (NO) release [3, [5], whereas mental arithmetic stimulates release of neural NO [6]. Thus, NO-mediated stimulation of soluble guanylate cyclase in vascular smooth muscle cells may have been involved in the vasodilator action of both intra-arterial saline infusion and mental arithmetic. A specific interaction at this final common path could have occurred, resulting in less arithmetic-evoked vasodilation in the infused arm as compared with the control arm.

Apart from bias, expression of results in terms of FBF ratio without adding an analysis of separate responses in the two forearms will leave possible systemic actions of local interventions undetected. This is illustrated in the paper by Petrie et al. (Figure 2 of their paper) where unilateral exercise clearly resulted in vasoconstriction in the nonexercising forearm, an action that is probably mediated by reflex sympatho-excitation, resulting in vasoconstriction in nonexercising parts of the body [7]. A similar phenomenon has been observed during infusion of adenosine into the brachial artery at subsystemic doses [8, 9].

In conclusion, although FBF ratio has the advantage of its higher reproducibility, a separate analysis of the two forearms remains necessary when results from forearm studies are reported in order to prevent bias and to detect possible (reflex-mediated) systemic actions of locally infused substances.