Bridging the gap between basic research on microcirculation and clinical world: The translational marriage between engineering and medicine

1 Introduction

From a general point of view, the acquisition of knowledge is the objective of basic research regardless any potential application to a practical need. Vice versa, applied research aims at directing the scientific knowledge towards practical applications. Findings of basic research can inform the design of applied research; applied research may be used to validate hypotheses of basic research. In the medical field, this interaction between basic and applied research is usually termed “bench-to-bedside”, which is associated with Translational Research.

The application of new scientific knowledge into clinical routine processes should be the aim of any biomedical research. This is the meaning of the term “Translational” when it is combined with “Research”: taken together, these words ultimately intend the improvement of the public health thanks to the advances of basic research.

Actually, several definitions of Translational Research are given in literature. The NIH stated that Translational Research includes two main areas: one is the process of applying discoveries generated during research in the laboratory and in preclinical studies, to the development of trials and studies in humans. The second concerns the research aimed at enhancing the adoption of best practices in the community [1].

Following the opinion of S.H. Woolf [2], the term “Translational Research” for many indicates the “bench-to-bedside” enterprise of harnessing knowledge from basic sciences to produce new drugs, devices, and treatment options; for others, Translational Research refers to translating research into practice, to be sure that new treatments and new knowledge reach the patients for whom they are intended. Other authors [3] argued that Translational Research fosters the multidirectional and multidisciplinary integration of basic research, patient-oriented research, and population-based research, with the long-term aim of improving the public health. In other words, Translational Research is part of a unidirectional continuum in which research findings are moved from the researcher’s bench to the patient’s bedside [3].

In 2015, the European Society for Translational Medicine (EUSTM) defined a new expression: Translational Medicine (TM). TM identifies an interdisciplinary branch of the biomedical field supported by three main pillars: benchside, bedside and community [4]. The goal of TM is to combine disciplines, resources, expertise, and techniques within these pillars to promote enhancements in prevention, diagnosis, and therapies. TM is a highly interdisciplinary field, whose primary goal is to coalesce assets of various natures within the individual pillars in order to improve the global healthcare system.

Beside any theoretical definition, the Director of the NIH, Dr E.A. Zerhouni, proposed a crucial question [5]: “what novel approaches can be developed that have the potential to be truly transforming for human health?”. The question is not trivial and the answer is hard. It is not easy to formulate an a priori prediction about the practical exploitation of any basic research.

Over the last decades, microcirculation has been receiving a growing interest from both basic research and clinical practice. Even though it plays a key role in the pathophysiology of many diseases, the translation of basic research on microcirculatory function into clinical practices is weak. From the one hand, the application of the Wavelet Analysis (WA) to Laser Doppler Fluxmetry (LDF) signals for investigating the microcirculatory function resulted in appreciable outcomes; from the other hand, the clinical community ignores, or at least underuses, this promising approach.

This paper identifies possible reasons of the poor translation of the WA to the bedside, and then encourages its dissemination.

2 Assessing the microcirculatory function by the WA

As it has been previously mentioned, an example of the gap between basic and applied research regards the application of the WA to the LDF signals for assessing cutaneous flowmotion waves. LDF signals do not only reflect skin microcirculation activities but also those coming from deeper tissues [6].

Flowmotion waves are provoked by the spontaneous and rhythmic dilation and constriction of the skin microvessels [7]. The frequency spectrum calculated from the flowmotion waves allows investigating those mechanisms that control and regulate the microcirculatory function (MF).

Microcirculation is where the exchange of substances between blood and tissues takes place; it comprises myriads of arterioles, capillaries and venules, closely coupled to the tissue perfused.

Microcirculation plays a key role in the pathophysiology of many diseases. Among others, the assessment of cutaneous hemodynamic has been considered an appropriate model for studying the MF [8–11] in hypertension [12–15], diabetes [16–18], and progressive systemic sclerosis [19].

In 2018, two studies have been published with regard to the assessment of MF: the first discusses the emerging application of semi-quantitative and quantitative nailfold capillaroscopy in systemic sclerosis [20], the second proposes the sidestream dark field imaging to assess preeclampsia microvascular dysfunction [21].

Over the years, the application of mathematical models that combine methods from both physics and engineering, has resulted in a continuing increase of knowledge and comprehension of the MF.

In particular, the WA was introduced in 1998: periodic oscillations with frequencies of around 1, 0.3, 0.1, and 0.04 Hz were demonstrated to represent the influence on cutaneous blood flow of heart beat, respiration, intrinsic myogenic activity, and the neurogenic factors, respectively [22, 23]. Afterwards, two further characteristic frequency intervals have been associated to the contributions of endothelial cells: 0.0095 – 0.021 Hz to NO dependent and 0.005 – 0.0095 Hz to NO independent activities (Table 1) [24].

The WA relies on a mathematical transformation that transfers the signal from the time domain (e.g., LDF signal) to the same signal in the frequency domain: in many cases, the information of interest is the spectrum of frequency content of the signal.

The WA overcomes many disadvantages of previous methods (Fourier Transform, FT; Short Time Fourier Transform, STFT). Actually, the WA optimizes the resolution in time and frequency: this is possible thanks to the optimal window used to scan the time points of the time-frequency plane [25, 26]. Several mother wavelets have been used thanks to their similarity with the shape of the waves to be analysed: the Morlet mother wavelet is considered the most suitable one for LDF signals, indeed [22, 23].

A family of dilated and compressed versions of the Morlet mother wavelet is created and shifted over the entire signal: it gets a coefficient for each pair translation-scale. The matrix of coefficients (scalogram) is constructed on three dimensions: time, frequency, amplitude of coefficients. Eventually, this matrix is averaged over time: the obtained vector represents the spectrum of the original signal (in arbitrary perfusion units, PU) (Fig. 1).

The intensity of each frequency component is expressed in amplitude (A) as absolute amplitude (AA) and relative amplitude (RA) (in average arbitrary units, PU), or energy (E) as absolute energy (AE) and relative energy (RE) (in average square arbitrary units, PU²).

OPEN IN VIEWER

Fig.1

From the original LDF signal (in arbitrary Perfusion Unit, PU), the matrix of coefficients (scalogram) is calculated by WA. The scalogram is finally averaged over time to get the spectrum of the signal.

A systematic literature review (from January 1990 to December 2017) discovered 98 studies on humans that applied the WA to the LDF signals [27]. Fifty-three studies have been performed on 892 healthy subjects; only 45 studies have been performed on 1679 patients in clinical applications.

As discussed in [27] with regard to healthy subjects only, 10 papers investigated the WA as a new method for assessing the MF; 9 analysed the effects of pressure on skin microcirculation and 7 the endothelial function at microcirculatory level; 7 papers assessed the cutaneous MF after thermal stimulation; 7 evaluated the ageing of microcirculation; 3 assessed the variation of the MF after muscular training; 3 studied the sympathetic activity; 2 considered tissue respiration. Moreover, 5 papers reported data on MF changes induced by infusion of saline solution, local variability of skin perfusion, post-occlusive reactive hyperaemia, application of magnetic field and hypoxia.

On the other hand, with regard to patients only, 9 papers investigated MF alterations in peripheral arterial disease (PAD), 8 after sympathectomy or nerve injuries, 6 in diabetes, 4 in melanoma and basal cell carcinoma, 3 in chronic pain, 2 in arterial hypertension, 2 in skin flap, 2 in general and local anaesthesia. Finally, 9 papers presented results of altered MF in lymphedema, myocardial infarction, heart failure, pressure skin ulcer, cerebrovascular disease, oestrogens effect, obesity, asthma, rheumatic disease.

The great advantage of the WA is the capacity to selectively identify and check the contributions of the cardiovascular activities influencing and controlling the MF: their origins are both intrinsic (myogenic, sympathetic, and endothelial) and extrinsic (cardio-respiratory). These activities are not distinguishable one from the other when assessed by means of any other methodology. In this sense, the WA is a powerful tool for extracting hidden information from the LDF signals.

The amount of papers published on both healthy people and patients undoubtedly demonstrates the great potential of the WA for the study of microcirculatory hemodynamic, and particularly for the assessment of endothelial function at microcirculatory level, under both physiological and pathological conditions. Nevertheless, the WA remains underused and it is still far from the clinical practice.

An interesting paper has been recently published to characterize the requirements for clinical microcirculation measurement techniques, highlighting the present barriers for translation into routine practice [28]. Authors stated that alterations in microcirculation are common in patients with shock, inflammation and sepsis; nevertheless, measurements of microcirculation have not yet entered clinical routine. International guidelines for the management of sepsis or cardiac anesthesia recommend specific goals targeting global hemodynamics but not microcirculation directly. This may be due to the lack of methods that fulfill the requirements necessary to be clinically acceptable.

Herewith, we try to list some (other) reasons of the poor translation of the WA to the bedside for the assessment of MF.

First, the method requires a theoretical expertise and some calculation resources that are neither immediate nor straightforward. As for many mathematical methods, WA effectiveness is strongly dependent on: 1) the robustness of the algorithms developed for its calculation; 2) their rigorous and affordable application. These conditions are standardly employed in a research laboratory, but they could be partly misapplied in the clinical practice. For example, signal acquisition has to be as longer as possible to properly check the lowest frequency components: at least 20 minutes are necessary to collect signals suitable for an affordable detection of endothelial cells activities. This long time interval sometimes impedes the reliable acquisition of LDF signals from patients with chronic ischemic pain [29]: the presence of many artefacts due to involuntary movements of the limbs impairs signals quality. Consequently, after LDF acquisition and before the WA, signals have to be processed for removing artefacts. At present, there is no commercial software making all these steps automatically. Therefore, another reason may be due to the lack of collaboration between medical units and research laboratories (of physics, engineering and bioengineering) that usually possess the theoretical expertise and the technical skills for the proper WA application.

Finally, the LDF has not yet been considered as a diagnostic tool for clinical routine purposes because of signal instability. LDF signals do not allow the definition of cut off thresholds to discriminate healthy subjects and patients.

On the one hand, the application of mathematical models that combine methods from engineering and physics has opened new perspectives for extending knowledge and comprehension of the MF; on the other hand, a gap does exist between the theoretical analysis of the MF and its clinical exploitation.

We believe that the first step to bridge the gap is the creation of a “common language” among basic researchers and clinicians. It will help sharing a common knowledge and, thus, it will allow connecting two separate communities: the community of mathematicians, physicists and bioengineers, and the community of clinicians, physiologists and vascular biologists. Their collaboration will result in a deeper understanding of the mechanisms that modulate and control blood flow oscillations in the microvasculature, under both physiological and pathological conditions. Their collaboration will ensure the necessary continuum from basic research to public health improvements, which only can accelerate and promote research translation for the creation of novel clinical strategies.

The establishment of a common language will allow more effective reciprocal dissemination actions. Basic researchers have to design and optimize the experimental procedures for extracting hidden information from the LDF signals, taking into consideration the limitations imposed by the clinical application; clinicians have to understand the theoretical fundamentals of the method to properly apply it for clinical purposes.

Actually, the translation will be completed once the consensus of the scientific community is obtained and operative guidelines are ascertained. In this process other stakeholders might be involved: international health care organisations will play a significant role in improving the connection between the findings of basic research and the clinical practice.

Going back to Dr Zerhouni’s question, we can conclude that the WA applied to the LDF signals has the potential to be usefully exploited for improving human health.