Abstract
Introduction
Tissue microcirculation and oxygenation plays a major role in the pathogenesis of diverse diseases [1–12] like infections, inflammation or wound healing [1–13]. Consequently measurement of tissue microcirculation or oxygenation awake interest for the follow-up of diverse operative procedures [14–16] in order to ensure success of the therapy.
However the majority of monitoring techniques did not succeed to meet all criteria’s of an ideal postoperative monitoring system. Positron emission topography, for instance, enables the detection of hypoxic areas. However this technique has low spatial resolution, is expensive and time-consuming [17]. Implantable 20MHz ultrasonic probes [18] are able to monitor the blood flow continuously. However, false negative values due to probe dislocation, vascular complications after removing the probe and potential infections prevented a widespread application of the implantable Doppler probe for postoperative monitoring.
Compared to current monitoring techniques assessment of tissue oxygenation with luminescence imaging seems to be an appropriate system for postoperative monitoring. Luminescence imaging meets most of the following demands of an ideal postoperative monitoring: it is harmless for the patient, fast, reliable and is easy to operate especially for inexperienced personal. Additionally promising results could be published over the last years [14, 19]. Compared to the current gold standard for transcutaneous oxygen measurement, the Clarke electrode, luminescence imaging even delivers more precise results. However hyperthermia and long measurement intervals had been necessary during the measuring process, which made this technique inappropriate for clinical use.
For this reason, we tried to assess tissue viability measuring transepidermal oxygen flux instead of the absolute oxygen partial pressure with luminescence imaging. The present study is the first application of oxygen flux imaging in clinical day-work assessing wound healing after fractures close to the hip joint.
Material and methods
Transepidermal oxygen flux
Human skin continuously takes up O2 from the atmosphere and eliminates carbon dioxide. This so-called skin respiration is essential for the oxygen supply of the upper dermis and the epidermis. 80–100 ml O2 per m2 were absorbed and 90–120 ml carbon dioxide per m2 were eliminated over the skin per hour [20]. Stücker et al. could demonstrate that oxygen uptake is inversely proportional to perfusion of the upper dermis and epidermis. Further investigations of Stücker could proof that pathologic changes of microcirculation could be assessed with oxygen flux measurement using luminescence imaging [20]. Compared to the current gold standard for transcutaneous oxygen measurement, the Clark electrode, luminescence imaging seems to delivers more precise results [14, 19].
Ratiometric luminescence imaging (RLI)
The key element of the monitoring device is a sensor foil, which is directly applied to the skin surface. Oxygen and carbon dioxide exchange between the skin and the sensor foil is unrestricted possible, whereas gas exchange between the atmosphere and the sensor foil is impossible. Solved indicator and reference molecules are excited by a light source. The excited indicator molecules collide with relaxed oxygen molecules resulting in deactivating of the indicator and transferring the energy to the oxygen molecule. The excited oxygen molecules relax without emitting fluorescence radiation. Measurement of an indicator response in relation to a reference response is more precise and less susceptible to light reflection and an inhomogeneous indicator dye distribution in the sensor foil compared to the measurement of absolute fluorescence intensity.
Sensor foil
The luminescent optical sensor used in our study is a polymer-based sensor (Sensor Prototype, PreSens GmbH, Regensburg, Germany). The sensor consists of green luminescent reference dye and red luminescent indicator dye immobilized in a polymer matrix and a polyester foil which is prepared by spreading a liquid reference/indicator polymer cocktail onto a transparent polyester support. After heat treatment the solvent evaporates and the remaining sensor foils are 3–5μm thick. The polymer is permeable for oxygen while being impermeable to any liquid at the same time and serves as a solvent for the indicator/reference dye. However, the polyester support is impermeable to oxygen. The sensor foils are not toxic and the size of the sensor foil can be chosen indiscriminately. In the presented study, a size of approximately 4×3 cm was chosen.
Fluorescence microscope
The setup was similar to previous investigations of our group [14, 21] and is briefly summarized in the following:
A portable RLI microscope (VisiSens A1 Prototype, PreSens GmbH, Regensburg Germany) was used. The microscope consists of a color camera, an LED-based excitation light source, optical filters and a lens, which allows the read-out of the sensor area. Data’s are acquired with a 1.3 megapixel color chip resulting in more than 300, 000 independent sensing points (= pixel of the respective color channel) for the indicator/reference response. The sensor response contains an indicator signal (red channel of the RGB-chip) and a reference (the green channel RGB-chip) signal. The indicator signal is set in relation to the reference signal and one single picture of the oxygen sensor is generated (R = red/green) [16].
VisiSens Software
The VisiSens AnalytsCal1 Software assured control of the camera settings and was used so calculate and analyze the oxygen sensor signal. Oxygen flux of a defined region can be calculated in a color-coded picture of the sensor signal after determining a defined region of interest. Oxygen flux was calculated based on 10 individual measurements every 2 seconds resulting in a period of 20 seconds.
Reduced tissue perfusion lead to low oxygen tensions and therefore increased diffusion of oxygen out of the sensor foil. Consequently more indicator molecules are in excited state resulting in more luminescence radiation and steeper slopes of O2-Flux (Fig. 1). In contrast, normal or enhanced tissue perfusion results in high oxygen tensions. Transepidermal oxygen flux is reduced and consequently more energy is transferred to oxygen molecules. Therefore recorded luminescence is lower and leads to flat curve-shape (Fig. 2).
Reduced skin microcirculation reflects reduced skin oxygenation and increased transepidermal oxygen flux of 0.085.
Enhanced skin microcirculation reflects enhanced skin oxygenation and reduced transepidermal oxygen flux of 0.028.
The clinical study was conducted at the Department of Plastic and Reconstructive surgery of the University hospital of Regensburg from 2013 to 2014. 22 patients with fractures close to the hips were examined at the operation wound. The contralateral side served as reference and was similarly investigated. Transepidermal flux measurement was conducted at the first postoperative day. All patients showed normal wound healing. No infection, seroma or delayed wound healing was observed. Age of the patients ranged from 28 to 94 years (71.9±15). 7 male and 15 female patients haven been examined.
Sensor foil placement was in direct proximity to the surgical scar. Reference measurement was on the corresponding area at the contralateral side. Sensor foil adhesion was improved by small amount of sodium chloride. Dimension of the sensor foil was 4×3 cm.
All patients were given verbal and written information of the nature of the study, and signed informed consent was obtained prior to study start. The trial protocol was reviewed and given favorable opinion by the Regensburg University Ethics Committee. The study was conducted in full accordance with the Somerset (South Africa, 1996) amendment of the Declaration of Helsinki (1964).
Statistics
Data of the postoperative analysis are given as mean±standard error of the mean (SEM). A t-test was applied to analyze differences regarding the transepidermal oxygen flux of the different measurement areas. A p-value below 0.05 was considered significant and marked using asterisks within thegraphs.
Results
6 out of 22 patients had to be excluded from statistical analysis because of light reflection or sensor foil displacement.
Mean of the transepidermal flux of the remaining 16 patients at the operation site was higher than the contralateral side. The differing transepidermal flux corresponds to a lower microperfusion of the operation wound compared to the contralateral side. However results have not been significant: operation wound 0.084±0.021, contralateral side: 0.071±0.029; p = 0.700. The 25 percentile was 0.280 at the operation wound and 0.280 at the contralateral side. The 75 percentile was 0.877 at the operation side and 0.605 at the contralateral side (Fig. 3).
Box plot diagram of the transepidermal skin flux measurement of the operations area and the contralateral side.
Measurement of skin oxygen is of great interest in diverse fields of medicine. Different pathologies, e.g. infection, ischemia, cancer or chronic wounds show a characteristic oxygen distribution and skin oxygen tension [1, 22]. Additionally, complex operative procedures in order to reconstruct soft tissue defects require reliable monitoring of the microcirculation [23]. Monitoring of free flap transplantation by the use luminescence lifetime imaging showed promising results in previous studies [16]. However measurement of transcutaneous skin oxygen with luminescence lifetime imaging needs the induction of hyperthermia and long measurement periods. These disadvantages made this reliable technique unsuitable for the daily routine.
Aim of this study was to assess skin oxygenation for postoperative wound monitoring using transepidermal oxygen flux instead of measurement of oxygen partial pressure with ratiometric luminescence imaging.
Twenty-two patients were investigated after operative treatment of fractures close to the hips. The corresponding area of the contralateral side served as reference for the measurement. Data of six patients had to be excluded due to light reflection or displacement of the sensor foil during the measuring period. Modification of the handheld microscope was necessary and after the supplement of an adapter, that prevent incidence of light to the camera, light reflection even under daylight conditions were eliminated (Fig. 4). Sensor foil adhesion was improved using a small amount of sodium chloride between sensor foil and skin. Consequently data acquisition and calculation of the remaining sixteen patients succeeded uneventful. Skin flux imaging in close proximity to the operation scar showed higher oxygen flux compared to the reference. This effect reflects the trauma during surgery resulting in a reduced skin perfusion at the operation site.
Modification of the handheld microscope with an adapter in order to prevent light incidence during the measuring process.
Similar results were obtained by Chang et al., who investigated skin oxygen tension after mastectomy, vascular-, cardiac- and abdominal surgery [24]. Right after surgery Chang et al. observed, that oxygen tension was lower at the operation area compared to the results of the control group.
Additionally big individual differences were seen between the patients at the same measuring area. The wide interquartile range of the 25 and 75 percentile displays this observation. Babilas et al. [19] also demonstrated, that assessment of dynamic changes are more informative than the evaluation of absolute values because of individual variations. Caspary et al. also observed a very high variation of tcpO2 in relation to the local skin temperature and the individual patient. According to this investigations absolute values of tcpO2 seem not to be suitable for postoperative wound monitoring [25].
Oxygen flux imaging using ratiometric luminescence imaging was easy to learn, even for inexperienced personnel. After the elimination of systemic errors due to light reflection and sensor foil displacement, reliable results were obtained in all patients. Consequently oxygen flux measurement succeeded almost all demands on a ideal monitoring system described by Jones et al. [26]. As sufficient oxygen supply is not only essential for wound monitoring [22, 27] but also suited for the assessment of different pathologies [1, 13] this technique offers diverse scopes of application.
However further investigations especially in a greater population and under pathologic conditions have to be conducted to proof these promising results.