gms | German Medical Science

Wounds – and especially those located on the lower extremities – are by definition chronic when they fail to heal within a period of 4 to 6 weeks, and additionally show no tendency to do so. They present a growing problem for our civilisation, in particular the chronic wounds occurring in CVI patients, mixed arterial-venous ulcers included, with the latter accounting for up to 80% of all lower leg ulcers [1], whereby the diabetic foot syndrome and decubital ulcers are not included. It is therefore no surprise – especially with regard to oxygen – that a comprehensive and longstanding collection of scientific literature is available [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], which cannot be discussed in detail here. Indisputable in the literature is the finding that the occurrence of such wounds is accompanied by hypoxia (see for example the S3 guideline for ulcus crusis venosum from the German Society of Phlebology [12]). Also incontrovertible is that anoxic tissue does not persist as such, but becomes necrotic. It is also known that (chronic) wounds undergoing oxygen treatment heal more rapidly and effectively [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], although it should be remembered that oxygen can also be not just oxidatively toxic, but can even prove to be deadly. Oxygen is thus paradoxically a life-sustaining, but deadly poison: It all depends on the use of the right dosage (see Theophrastus Bombastus of Hohenheim, also known as Paracelsus).

Using biomolecular investigations it could be shown that oxygen not only acts as a substrate for cellular ATP synthesis, but that it can also form reactive molecules (reactive oxygen species (ROS)) such as peroxide and superoxide anions, and hydroxyradicals which are the key signalling substances for the inflammatory reaction in wound healing and for phagocytosis in leucocytes and macrophages (although when present in excess capable of causing tissue damage). Oxygen has also been found to have a strong antibiotic effect in wounds and hydroxylases require an oxygen partial pressure of approximately 70 mmHg in order for collagen synthesis to occur satisfactorily. Additionally, it should be noted that a physiologically regulated (still compensated) non-pathological hypoxia is capable of initiating and supporting important wound healing sub-processes such as the release of vascular endothelial growth factor (VEGF) which is involved in the vascularisation of skin tissue. As opposed to an acute wound, a typical chronic wound is characterised by an altered metabolism and modified cell reactions. Examples of these are: e.g. an increased amount of matrix metalloproteinases (MMP) found in the locality of the wound which is responsible for the intensified hydrolysation of cytokines and growth factors capable of promoting wound healing [13], e.g. an intensified degradation of the extracellular matrix [14], e.g. an inhibition of cell proliferation in wounds in which the exudate contains an increased amount of inflammatory cytokines [15], and where increased amounts of the aforementioned reactive oxygen species (ROS) are present [16]. This is referred to as cellular senescence [16], [17], [18], [19], [20], [21].

According to our thesis, most chronic wounds are the result of local oxygen deficiency, in which the cells at the base of the wound are only just capable of maintaining their structure, such that they are barely able to maintain the metabolic rate necessary for tissue conservation. As a result of hypoxia, the cells at the base of the wound lack metabolic energy, in particular in the form of oxygen (and subsequently adenosine triphosphate (ATP), but possibly also further necessary substrates) for tissue regeneration. The cells still however have the potential to regenerate and can – with an adequate substrate supply – immediately start regenerating. Due to the prevailing loss of tissue however, the wound base can only regenerate normal skin at an enormous energetic and metabolic expense. In order to enable the healing of a hypoxic chronic wound, the local hypoxia must be compensated by a substitution of oxygen, analogue to the natural supply of the skin either with an increase in the oxygen loading of the blood or from the exterior. For this purpose we have previously developed appropriate procedures [22], [23]. The regeneration and healing of the skin should constantly take place under normoxic conditions if possible, so that the skin regeneration and healing of the wound can be accomplished with high quality collagen and not with substandard, instable scar tissue which can degenerate at any moment, and again form a chronic wound (recidivation). In a wound in which the tissue hypoxia is not too pronounced, the available oxygen will be sufficient enough to regenerate the skin region following the healing of the ulcer.

The given dependence of wound healing on an adequate supply of oxygen also requires the determination of a reliable and meaningful oxygen status for the wound concerned so that, for example, it is possible to ascertain whether or not an oxygen substitution should be given, and if so, in which form it should preferably be given. For the determination of a clinically relevant tissue hypoxia, only parameters which provide a measure of the available oxygen (available for the mitochondria) are suitable. Therefore, for the characterisation of hypoxia only the (non-invasively measured) transcutaneous oxygen partial pressure (tcPO₂) comes into question as it is a direct measure of the tissue oxygen partial pressure (gPO₂) and in turn, the ultimate parameter for the driving force of the diffusive provision of the tissue cells (i.e., the mitochondria) with oxygen. Invasive methods for the determination of tissue pO₂ are not suitable for routine measurements; the oxygen saturation of haemoglobin, the oxygen content of blood and the blood perfusion of the skin are surrogate parameters, which do not reliably reflect the oxygen status (for example, in order to estimate the local partial oxygen pressure using the oxygen saturation of haemoglobin, an exact knowledge of the local oxygen binding curve for haemoglobin is required. Even less reliable is a deduction using the blood perfusion; to the contrary, in CVI wounds, hypoxic tcPO₂ values and at the same time blood flow levels which are way above the norm are found using the laser Doppler method [24]).

In the past, tcPO₂ measurements have, for example, been used to determine the amputation level in order to be certain that the operation wound would heal [25], [26], [27]. In the literature, information can be found concerning tcPO₂ measurements in the proximity of wounds, although exact details are usually lacking. Peri-ulceral tcPO₂ measurements close to the edge of chronic wounds were obtained by Jünger et al. [28]; they found coefficients of variation (quotient of the sample error and mean) of up to 80%. Such pronounced variance can only be due to biological factors, since the intraindividual reproducibility in the form of the coefficient of variation of repeated serial peri-ulceral tcPO₂ values at the same location in a chronic wound is approximately 10% [23]. On the other hand however, such measurements at different locations around a chronic wound produced, for example, values of 9.9, 2.8 and 19.5 mmHg, which resulted in inhomogeneous healing [23]. This also suggests extreme inhomogeneity in the peri-ulceral oxygen partial pressure of individual chronic wounds, whereby 14 tcPO₂ values were sequentially measured with a probe in just one wound within 3 weeks, and in part repeatedly reproduced (sequentially determined oxygen inhomogeneity). Due to this inhomogeneity, single peri-ulceral tcPO₂ values from a chronic wound do not provide reliable and representative information concerning the oxygen status of the entire wound. Instead, a method is required which, for example, provides several (and preferably simultaneous) tcPO₂ values, while at the same time being suitable for diagnostic, therapy selection, prognosis, observation and recidivation prophylaxis purposes.

The underlying condition of patients with chronic wounds was determined using the ankle brachial pressure index (ABPI) of the (ultrasonographically measured) systolic blood pressure, with photosplethysmography (venous refill time after a muscle pump test “Quantitative Photoplethysmograph Vasoquant VQ1000 D-PPG”, ELCAT, Wolfratshausen), with duplex sonography (venous backflow time, ultrasound diagnostic system “Nemio SSA-550A”, Toshiba, Neuss) and with the (fasting) blood glucose level.

The volunteer and patient collective was made up of:

1.

21 healthy volunteers, in which tcPO₂-measurements were carried out peri-malleolarly around the right inner ankle. The mean age was 40.6 (±13.7) years (mean ± standard deviation) with a range from 22 to 72 years.

2.

17 Patients with – in most cases pure – chronic CVI wounds in the ankle region. Data were collected retrospectively from the corresponding documentation on wound treatments. The mean age of the patients was 66.3 (±14.3) years with a range from 42 to 82 years.

The measurement system “tcpO₂ Monitor Tina TCM 400” (Radiometer, Willich, Germany) was used for the simultaneous determination of peri-ulceral tcPO₂ values with 4 electrodes. This was sufficient for measurements in all the smaller wounds with a maximum diameter of approximately 8 cm, for larger wounds, use of the 6 electrode system might be preferable in order to obtain a more detailed assessment of oxygen inhomogeneity.

Prior to each measurement, the electrodes – which were heated to 44°C – were calibrated according to the manufacturer’s recommendations. It is also advisable to ensure that the electrodes are sufficiently stable before the wound measurement (which lasts approximately one hour): After calibration, the electrode is taken out of the storage case and kept in air contact for the measurement duration of about an hour. During this time the device registers values in the curve mode, such that the electrode signals are visible. After one hour, the drift should not be more than 10 mmHg (baseline measurement). For the actual measurement, the 4 electrodes are calibrated again and then remain in air contact during which the respective air output signal is recorded (see also Figure 1 [Fig. 1]). The skin area is then degreased using alcohol and the electrodes affixed to the peri-ulceral skin as close as possible to the wound margin (for the device used this distance was approximately 15 mm), with the O₂-sensitive central measurement site of the electrode located as close as possible to the wound margin and preferably accordingly to a set pattern around the wound in the positions “North” (proximal; electrode 1), “East” (right; electrode 2), “South” (distal; electrode 3), “West” (left; electrode 4) as can be seen in Figure 2 [Fig. 2]. Once the electrodes have been applied appropriately, the signal decreases rapidly, passes through a minimum, and after about 15 minutes reaches a plateau, from which the measurement values are determined.

The patient (or volunteer) then inhales pure oxygen through a mask (with a flow of somewhat more than 10 L/min, i.e., in abundance) for about 15 minutes, whereby the signal usually then changes to a new value. After completion of this inhalation the signal rapidly returns to the first plateau, thus allowing a control of the measurement value. The electrodes are then removed from the wound, and for control purposes left exposed to room air for a few minutes, after which the starting air output signal (or a value within 10 mmHg of it) should be registered. The measurement is then complete. Figure 1 [Fig. 1] shows the simultaneous registration on the ankle of a healthy volunteer and on a typical CVI wound. The device is in curve mode, for more details of the measurement technique see [29].

For the measurement in the ankle of a healthy volunteer, the values 64, 76, 69 and 50 mmHg were obtained with the 4 electrodes and for the chronic CVI wounds the values were 31, 5, 3 and 2 mmHg.

During O₂ inhalation, the registration of values in the healthy volunteer showed pronounced and synchronised tcPO₂ oscillations, which were (typically) less pronounced in the chronic wounds.

Using the new technique, oxygen status information which is as exact and reliable as possible should be obtained. For this reason, 4 electrodes are used simultaneously, with the use of 6 electrodes being recommended for larger wounds. It is certainly not meaningful to use the mean value of all of the tcPO₂ values as a hypoxia parameter since, in cases in which an enormous inhomogeneity exists (see the example given above) a partial severe hypoxia – which would be a determining factor in the healing of the chronic wound – would not be adequately taken into consideration with such a parameter. At the same time however, the use of the lowest tcPO₂ value as a single value for the characterisation of the whole wound would possibly be too unreliable (since a technical or measurement error from a single electrode would falsify the measured oxygen status for the complete wound.

As a result, the parameter which has been used here to describe the oxygen status of a chronic wound (K-PO₂) is the arithmetic mean of the two lowest peri-ulceral tcPO₂ values measured (as a measure of the oxygen tension of the lower values (with 4 electrodes, half of it).

The K-PO₂-value for the healthy volunteer mentioned above was 58.5 mmHg, whereas that of the abovementioned patient was 2.5 mmHg.

Since a large inhomogeneity in the peri-ulceral tcPO₂ values around an individual chronic CVI wound is usually found [28], it appears to be appropriate that a new respective characteristic should be introduced. For this purpose we have defined the oxygen inhomogeneity of a chronic wound (I-PO₂) as the variation coefficient of the tcPO₂ (quotient of the sampling error and the arithmetic mean of 4 peri-ulceral values).

For the abovementioned healthy volunteer the oxygen inhomogeneity was 17%, whereas for the patient it was 135%.

Using the new oxygen parameter (K-PO₂), the hypoxia within a chronic wound can now be defined and characterised. In the relevant scientific literature the general statement can be found declaring that a tissue with a tcPO₂ value of less than 40 mmHg is hypoxic, since it is considered to be no longer capable of regeneration (see for example [29], [30], [31]). We have therefore defined a chronic wound as being hypoxic when the K-PO₂ is less than 40 mmHg.

This rough definition is however not adequate for a detailed targeted therapy, since a chronic wound with a K-PO₂ of 2.5 mmHg (see above) requires a completely different treatment modality to that of a wound with a value of, e.g., 33 mmHg. For this reason – and in order to be able to verbalise different degrees of hypoxia – a grading of the extent of hypoxia is necessary, which has been arbitrarily undertaken here with the help of the new oxygen characteristic K-PO₂, as seen in Table 1 [Tab. 1].

Initially, tcPO₂ measurements were carried out peri-malleolarly around the right inner ankle in 21 healthy volunteers as described. The region was selected as the ankle regions are predilection sites for chronic CVI wounds. Table 2 [Tab. 2] shows the data and measurement results for the individual volunteers as well as the evaluation in terms of the oxygen characteristic (K-PO₂), the oxygen inhomogeneity (I-PO₂) and the oscillations.

None of the healthy volunteers were found to be hypoxic on the right inner ankle according to the new definition, i.e., the parameter K-PO₂ was, in all cases greater than 40 mmHg, with a mean value of 63.0 and a standard deviation of 10.4 mmHg. The range was substantial with values from 40.5 to 77.5 mmHg being measured. The mean oxygen inhomogeneity and its distribution (standard deviation) was 14.1 (±6.2)%. Practically all of the healthy volunteers exhibited synchronised oscillations in the tcPO₂ signal during O₂ inhalation.

A summary of the data and results in the 17 chronic (predominately purely venous) wounds can be seen in Table 3 [Tab. 3].

The mean oxygen characteristic (K-PO₂) was 17.9 (±12.7) mmHg, whereby values over the total hypoxia range from 0 to 40 mmHg are included. None of the K-PO₂ values is above 40 mmHg, so that all values, by definition, are hypoxic. Only the hypoxia grade of wound No. 14 lies in the narrow, compensated, physiologically regulated hypoxia range, and only this particular wound showed no pathological oxygen deficiency.

In Figure 3 [Fig. 3] a comparison of the frequency distributions of all of the ankle and peri-ulceral measurement values can be seen. The two distributions are entirely different. Whereas the ankle values show a distribution maximum at 75 mmHg, with a range from 40 to 90 mmHg and a distribution which is slightly sloping to the left, the peri-ulceral values show no distribution maximum, but instead show only a distribution plateau of approximately 20% (using in each case the same class widths of 10 mmHg) which ranges from 0 to 35 mmHg, with a maximum value of 60 mmHg. Thus the pathological peri-ulceral tcPO₂ values are distributed evenly widely and are therefore extremely inhomogeneous.

The (4) tcPO₂ values from single individual chronic CVI wounds, i.e., intraindividual values, are characterised by an enormous inhomogeneity, which is quantitatively captured here in the value for the coefficient of variation (I-PO₂). Whereas the mean of this coefficient for the skin of the ankle was only 14.1% and remained practically constant, in the case of the CVI wounds the mean value was 54.6% which is more than 3 times greater (note that simultaneously determined O₂ inhomogeneities are being considered here). The inhomogeneity (I-PO₂) is however itself not constant, but as can be seen in Figure 4 [Fig. 4], is a characteristic function of the oxygen characteristic (K-PO₂) which rises exponentially with increasing hypoxia and which in extreme hypoxia (K-PO₂ close to zero) shows an approximately 10-fold increase, with values of about 140%. For values of less than 40 mmHg, the K-PO₂ characterises the degree of hypoxia (see also Table 1 [Tab. 1]).

Figure 4 [Fig. 4] shows the regions of normoxia and hypoxia, together with the relatively small physiologically regulated (compensated) region, in comparison to the large pathological region. Whereas the extent of oxygen inhomogeneity in the normoxic region is only small and remains almost constant, in the hypoxia region it rises exponentially with a declining oxygen parameter. However, the physiologically regulated hypoxia region is not affected; instead the exponential increase only starts to appear in the pathologically hypoxic region. The increase in inhomogeneity occurs – as can be seen in Table 3 [Tab. 3] and the distribution in Figure 3 [Fig. 3] – one-sidedly through the development of hypoxic peri-ulceral skin regions and not due to hyperoxic areas, as exemplified by the “left-skewness” of the distribution of the peri-ulceral measurement values with a plateau.

To begin with, a discussion on whether peri-ulceral tcPO₂ measurements can be considered to be suitable for the assessment of the extent of hypoxia in chronic wounds is necessary. We consider such measurements to indeed be suitable since the chronic wounds and the peri-ulceral skin must be seen to be a patho-functional unit since the (necrotic) chronic wound can only arise in degenerated skin, even with regard to the temporal order of events: The decrease in the tcPO₂ values precedes the formation of the wound [22]. This also becomes clear in Figure 5 [Fig. 5] where extreme hypoxia – though not yet a chronic wound – can be seen, although the clinical signs of hypoxia, namely depigmentation (atrophie blanche) together with brown pigmentation (melanin) and haemosiderin deposition (purpura jaune d’ocre) are evident. When part of the tissue becomes necrotic, an ulcer forms and the remaining (non-necrotic) tissue becomes the peri-ulceral surrounding. Subsequently, the immediate peri-ulceral skin, as well as the sub-ulceral tissue, are both still equally only just capable of survival due to the provision of oxygen to the tissue. Furthermore, a tcPO₂ gradient then exists from the most distant tissue parts (with only minor hypoxia) to the ulcer, and in particular, from the next measurable peri-ulceral skin area (located at least 15 mm from the rim of the ulcer, see above) to the immediate sub-ulceral (intra-ulceral) tissue. According to this approach it must then be assumed that the peri-ulceral tcPO₂ values, as a measure of the intra-ulceral tcPO₂ values will have a tendency to be incorrectly large such that the extent of hypoxia in the wound will be underestimated (see below). During healing, the sequence reverses: Primarily, the peri-ulceral skin must regenerate before the wound can close over the base of this regeneration. These two processes will of course overlap.

A further fundamental question is whether the epicutaneous tcPO₂ measurements characterise the cutis itself in terms of oxygen. For this purpose, Roszinski and Schmeller [32] carried out comparative intracutaneous measurements, in each case at the same place on the skin using needle electrodes (icPO₂) placed at a depth of up to 2 mm, in intact skin of the lower leg and in the peri-ulceral region. The authors found significant differences in the tcPO₂ measurements (at 44°C) between intact and CVI skin with values of 59 (±12) and 5 (±6) mmHg, respectively (mean ± standard deviation); in the case of icPO₂ measurements the values were 51 (±9) and 22 (±10) mmHg, respectively, which were also significantly different. During insertion of the needle in the skin by up to 8 mm, the authors found – in contrast to intact skin – PO₂ differences of up to 50 mmHg. The intra-individual oxygen inhomogeneity is therefore not limited to the epidermis, but is also evident intracutaneously. Altogether, the icPO₂ results confirm the tcPO₂ results, although whereas the comparative values for the intact skin are in agreement, the tcPO₂ values for the peri-ulceral skin are significantly smaller than the icPO₂ values (see above). This indicates that the supply of oxygen to the epithelium of the peri-ulceral skin is poorer than that of the intracutaneous tissue, whereas this is not the case in intact skin. This conclusion becomes even more acceptable when the fact that the authors were unable to find a correlation between the two measurement procedures in either intact or peri-ulceral skin is considered.

Since both of the oxygen parameters for chronic wounds (K-PO₂ und I-PO₂) have been newly defined here, only sporadic observations concerning their clinical relevance are available. None of the K-PO₂ values from the 17 chronic wounds investigated were found to be above 40 mmHg and therefore they are all – by definition – hypoxic. Only the hypoxia grade of wound number 14 lies in the narrow compensatory physiological-regulatory hypoxia range and only this particular wound shows no signs of a pathological oxygen deficiency. However, it should be taken into consideration that the determined hypoxia value (K-PO₂) tends to be greater than the most hypoxic skin value around the wound and that this is so for two reasons: (1) It is very improbable that the definitional stipulation (12, 15, 18, 21 o’clock) will exactly “hit” the two lowest values, and (2) the measurement is peri-ulceral and not intra-ulceral, although due to necrosis, the greatest oxygen deficiency occurs intra-ulcerally (see above). The other 16 CVI wounds (i.e., 94% of the wounds) are decompensated and pathologically hypoxic and thus require oxygen. Therefore, there is a compelling need for a therapeutic oxygen substitution.

The presented results indicate – in terms of properties crucial for the healing of chronic wounds (namely hypoxia) – that chronic wounds are extremely inhomogeneous in two respects: (1) the grade of hypoxia in different wounds (inter-individual) is drastically diverse, and (2) peri-ulceral tcPO₂ values within a few centimetres of each other in the same wound are extremely different to one another. The strong inhomogeneity justifies our use of arbitrary hypoxia grading. Tissue hypoxia may possibly always be coupled with (increased) oxygen inhomogeneity. The enormous oxygen inhomogeneity in the pathological hypoxia region also shows why it is not possible to reliably characterise the oxygen status of a chronic CVI wound using a single peri-ulceral tcPO₂ measurement. It is instead necessary to assess the peri-ulceral wound margin with as many electrodes as possible in order to identify hypoxic skin areas. Seen from a methodological aspect, it would be preferable if the edge of the wound were to be monitored continuously, to ensure that the most hypoxic areas are exactly located.

From the aforementioned oxygen inhomogeneity found in the sequential measurement of a single chronic wound [23] it becomes clear that this condition remains stable for a period of at least several weeks, and only slowly decreases during the healing process. In the case of a chronic leg wound of arterial origin with pronounced lipodermatosclerosis, we were able to show that during the healing of the wound the characteristic K-PO₂ increased and even normalised, while at the same time the variance of the measurement values decreased concomitantly with the magnitude of the I-PO₂ [22].

The results presented here have been confirmed by similar measurements performed by Falanga et al. [33]. The authors carried out measurements in 14 venous ulcerations, as well as in 5 non-ulcerated lower legs in CVI patients and 6 healthy lower legs. All measurement values could be evaluated using our new procedure. The mean oxygen characteristic (K-PO₂) in the ulcers was 12.0 mmHg, with a range from 1.5 to 36.5 mmHg, indicating that all of the chronic wounds were by definition hypoxic. The mean oxygen inhomogeneity value (I-PO₂) was found to be 59.6%. In the non-ulcerated lower legs with CVI, the mean K-PO₂ was 40.6 mmHg, with values ranging from 31.5 to 47.5 mmHg, i.e., the oxygen supply was only marginal, with some skin areas being hypoxic, while others – according to our definition – were normoxic. The mean oxygen inhomogeneity with a value of 18.7% was distinctly lower than the corresponding values in chronic wounds of 59.6%. In the healthy lower legs, the mean oxygen parameter (K-PO₂) was 54.8 mmHg with values ranging from 47.5 to 60.5 mmHg, i.e., no skin areas were found to be hypoxic. The oxygen inhomogeneity amounted to only 11%.

Ogrin et al. [34] also carried out peri-ulceral tcPO₂ measurements at 44°C in 13 chronic wounds. They found low values which were highly significant (p=0.008) in comparison to measurements in non-ulcerated lower legs. They did this by conducting mirror-image measurements on the contralateral leg of those patients who had undergone measurements in their ulcers.

The data in Table 2 [Tab. 2] and 3 [Tab. 3] also show that synchronised tcPO₂ oscillations (with frequencies around one per minute) were evident in ankle measurements performed during the oxygen inhalation challenge, but that such oscillations were not seen in chronic wounds. The lack of oscillations appears to be a characteristic of chronic wounds. It has been known for some time that PO₂ oscillations with this frequencies occur in the blood and measured transcutaneously of arterial vessels [35], [36]; they are attributed to vasomotion and are the result of a currently increased regulatory vascular constriction occurring as a response to local hyperoxia. If this interpretation is correct, then the physiological constrictive vascular hyperoxia reaction of the peri-ulceral skin and thus also the arterial vascular regulation of the peri-ulceral skin are defect. From another point of view, the synchronised tcPO₂ oscillations are characteristic of an intact arterial system and may, in general, be used for testing and evaluation purposes.

An arterial defect in the peri-ulceral skin can also explain why the tcPO₂ increases during the oxygen inhalation challenge were not significant in the chronic CVI wounds. The wounds partly showed increases in the tcPO₂ which were greater than those found in the skin of the ankle, whereas other wounds exhibited values which were lower. The higher values can be explained by a defect in the autonomic vessel regulation and the lower values can occur as a result of atherosclerosis or even a combination of both of the mentioned factors, according to the type and extent of the pathological skin and vessel alterations present.

From these considerations it would appear that the hypoxic tissue areas with a defect hyperoxia regulation during oxygen inhalation are given priority in terms of oxygen supply, an effect that is desirable and especially advantageous for regeneration, whereas the vascular regulatory intact regions remain protected from oxidative oxygen damage.

The use of the oxygen inhalation challenge (with inhalation of pure oxygen) during the measurement allows an assessment of the extent to which the measured skin regions around the chronic wounds can still be reached with systemically applied (inhaled) oxygen. In healthy subjects, an increase in the tcPO₂ was always seen during the oxygen inhalation. This indicates that the hyperoxia-induced vasoconstriction in the intact skin does not completely hinder the delivery of oxygen. In patients however, an increase in the measured peri-ulceral transcutaneous oxygen tension is a prerequisite if an oxygen inhalation is to be used for a local substitution of oxygen. The absence of such an increase suggests that a therapeutic oxygen inhalation would be futile.

The authors thank Messrs E. Martinez Mena, V. Zablah Larranga und R. Kiparski (Mexiko) for the provision of the facilities and allowing the use of the equipment necessary for carrying out measurements in the chronic wounds.

The authors declare that they have no competing interests.