Assessment of Microcirculation of an Axial
`Skin Flap Using Indocyanine Green
`Fluorescence Angiography
`
`Serdar Eren, M.D., Albert Rubben, M.D., Rainer Erein, M.D., Grahame Larkin, M.D.,
`and Rolf Hettich, MJD.
`Cologne, Germany
`
`In many cases the complexities of skin-flap microcir­
`culation are difficult to assess despite all the subjective and
`objective examination techniques available today. Ade­
`quate microcirculation is essential for tissue viability, so
`any method employed for studying microcirculation
`should provide as accurate an assessment of the prevailing
`conditions as possible. Of all the clinical methods, the
`fluorescence technique using the dye sodium fluorescein
`has so far provided the most reliable results. However, the
`pharmacokinetic properties of this tracer have prevented
`the technique from becoming established in clinical prac­
`tice. The fluorescent dye indocyanine green (Cardio
`Green), on the other hand, has far more favorable phar­
`macokinetics.
`In an experimental animal model, the fluorescence
`technique using indocyanine green (indocyanine green
`angiography, ICGA) was used to study postoperative
`changes in the microcirculation of a skin flap. On the day
`of operation, indocyanine green angiography revealed a
`state of hemodynamic imbalance for which the organism
`was able to compensate in the postoperative phase with
`the aid of humoral, physical, and metabolic factors. With
`indocyanine green angiography it was possible to quantify
`objectively the new hemodynamic equilibrium. Basically,
`microcirculation may be quantified in temporal and spa­
`tial terms. The significant objectivity of indocyanine green
`angiography and short intervals between each examina­
`tion favor its possible and meaningful use in clinical prac­
`tice and give cause for continuing studies. (Plast. Recon-
`sir. Surg. 96: 1636, 1995.)
`
`and, on the other, adapts itself to local condi­
`tions and requirements. Adequate microcircu­
`lation is, however, essential for tissue viability
`and functionality. Autoregulatory mechanisms
`over time and space control the basic demands
`of the skin. These mechanisms include first and
`foremost the neural regulation of the vascular
`system. In addition, humoral, metabolic, and
`physical factors exert their influence on the
`microcirculation.
`Similarly, adequate microcirculation is essen­
`tial for the viability of a skin flap. With regard
`to the physiologic conditions of skin perfusion,
`even the dissection of the flap has a significant
`effect on its microcirculation, as well as on its
`autoregulatory mechanisms, leading to a hemo­
`dynamic imbalance in its circulatory system.
`The organism tries to compensate for this by
`humoral, metabolic, and physical mechanisms.
`The otherwise dominant neural regulation of
`the microcirculation is disrupted by flap dissec­
`tion. If the organism is not able to restore an
`adequate microcirculation, then necrosis of the
`skin flap will occur. Skin perfusion is complex
`even under physiologic conditions; with a skin
`flap, one is confronted with even more difficulty
`when assessing microcirculation.
`McCarthy1 describes various subjective and
`The microcirculation of the skin is a very
`objective methods of examination to assess mi­
`complex chain of events which, even under
`crocirculation, the most common being the
`physiologic conditions, is characterized by its
`clinical examination. Inspection of skin color
`heterogeneous behavior. Of the total blood
`and capillary blanching on pressure gives clues
`flow through the skin, functional circulation
`to the current state of microcirculation. Mon­
`plays a greater role than nutritive circulation.
`itoring skin temperature also can be very help­
`The heterogeneity of resting perfusion is, on
`ful. Stabbing with a cannula or a no. 11 scalpel
`the one hand, dependent on anatomic factors
`From the Clinic of Plastic and Hand Surgery at St. Agatha Hospital and the Clinic for Burns and Plastic and Reconstructive Surgery and the
`Dermatological Clinic at the RWTH Aachen. Received for publication October 26, 1993.
`1636
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 1 of 14
`
`

`

`1637
`Vol. 96, No, 7 / MICROCIRCULATION OF AN AXIAL SKIN FLAP
`blade to induce capillary bleeding is another
`ripheral vascular diseases. They used the dye
`effective test forjudging flap circulation. Inter­
`sodium fluorescein to gain an impression of the
`pretation of these examination results, how­
`conditions of microcirculation from the inten­
`ever, requires a sufficient degree of clinical ex­
`sity, rate, and homogeneity of the appearance
`perience. On the whole, these tests are easy to
`of fluorescence in the tissue.
`apply yet must be considered unreliable.
`Sodium fluorescein is a water-soluble fluores­
`These purely subjective tests are contrasted by
`cent dye with a molecular weight of 376. In
`objective procedures for assessing microcircula­
`plasma and whole blood it displays an absorp­
`tion that are employed in clinical practice and
`tion maximum at 495 nm and an emission max­
`research with varying degrees of success and sig­
`imum at 515 nm.13 Approximately 50 percent of
`nificance. These methods include measuring
`systemically applied sodium fluorescein is
`transcutaneous PQ^ (tcP0), vital capillaroscopy,
`bound primarily to albumin, as well as to the
`photoplethysmography, laser Doppler flowmetry,
`surface of erythrocyte membranes.14,15 Fluores­
`thermography, isotope clearance, dermoflu-
`cein is freely dissolved in plasma and, as a result
`orometry and dermofluorography, and measure­
`of its low molecular weight, can diffuse into the
`ments using radioactively labeled corpuscular
`interstitium by a transcapillary route.2
`blood components.
`It was Myers16 who in 1962 used sodium flu­
`Measuring tcP0g, a selective technique, has so
`orescein in operations to predict skin necrosis
`far not produced the expected results with re­
`by differentiating between fluorescent and non-
`gard to the complexities of skin circulation, and
`fluorescent areas. By using the dermofluorom-
`because of its susceptibility to interference, it is
`eter, Silverman et al.17 quantified the uptake of
`now rarely used. Vital capillaroscopy is an ex­
`sodium fluorescein in the skin and correlated it
`cellent method for recording local mechanisms
`with flap viability. Since then, various studies
`of microcirculation (flow distribution, erythro­
`have shown good correlations of the fluores­
`cyte flow rate, capillary diameter, etc.). Further­
`cence technique with blood flow and the via­
`more, qualities of transcapillary and interstitial
`bility of tissues.18-26
`diffusion can be determined by injecting a flu­
`Although dermofluorometry is commonly
`orescent dye.2,3 Despite modifications of the
`used by various authors, it only allows a punctate
`method, photoplethysmography4 has so far not
`registration of tissue fluorescence. This is why
`produced unequivocal results in the assessment
`Lund employed a rapid-sequence photo­
`of skin-flap circulation, so this technique is now
`graphic recording of dye distribution to gain a
`seldom used. Great expectations were placed
`spatial analysis of skin perfusion.27,28 This tech­
`on the laser Doppler flowmeter. As a noninva-
`nique was modified further by Rieger and
`sive method, it is simple and can be repeated
`Schefiler29 using a digital image processing sys­
`several times in succession. Serial monitoring
`tem. However, the pharmacokinetic properties
`allows registration of local autoregulatory fluc­
`of sodium fluorescein have until now prevented
`tuations of skin perfusion. Plastic surgery in
`this method from becoming further established
`particular can look back on a number of inter­
`in clinical practice. Its routine use for assessing
`esting studies.5-9 The disadvantages of this pro­
`blood circulation of skin or a skin flap was un­
`cedure are its susceptibility to artifacts due to
`successful, not least because of the long inter­
`movement and its inability to provide an overall
`vals (7 to 8 hours) between examinations.
`picture of the microcirculation. Extensive data
`The vital dye indocyanine green (ICG) was
`are provided by thermography, which, despite
`first introduced into clinical medicine by Fox et
`its good approach, however, is not particularly
`al.30,31 in 1957. Indocyanine green is used pri­
`popular. The most exact study results have so
`marily in hepatology as a liver function test32,33
`far been provided by the clearance method us­
`and in cardiologic diagnostics.34"37 In 1973,
`ing Xe, H, and Te and by the process using
`Flower and Hochheimer38,39 introduced it into
`radioactively labeled corpuscular blood compo­
`the fluorescence technique. They used the dye
`nents.10,11 But for reasons of time and cost, these
`for fluorescence angiography of the choroidea.
`tests are used principally in research.
`Indocyanine green has a molecular weight of
`The fluorescence technique currently pro­
`775 and is almost completely bound to plasma
`vides the most accurate information on the mi-
`proteins following intravenous application,
`crocirculatory state of skin. It was first used by
`with alpharlipoproteins and albumin being the
`Lange and Boyd12 to study circulation time, cap­
`principal binding partners.40,41 Tight binding to
`illary permeability, and tissue circulation in pe­
`plasma proteins guarantees that the fluorescent
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 2 of 14
`
`

`

`1638
`dye will remain intravasal. During its elimina­
`tion from the blood, indocyanine green under­
`goes biphasic plasma clearance. In the initial
`phase, tl/2 amounts to 3 to 4 minutes at a dose
`of 0.5 mg/kg of body weight, with a t1/2 of 66
`minutes in the second phase.42 More than 90
`percent of the applied dye is eliminated in the
`first phase. These pharmacokinetic properties
`offer a considerable advantage over the behav­
`ior of the dye used hitherto. Further studies on
`the choroidea,43 as well as those using fluores­
`cence videomicroscopy,3’44 also have produced
`good results with indocyanine green, so it would
`seem reasonable to use this dye in a broader
`model using skin or a skin flap. The rapid elim­
`ination of the dye also makes the possibility of
`a time-oriented analysis of skin perfusion more
`likely. The present animal study provides data
`on the microcirculatory state of an axial-pattern
`flap by means of a computer-assisted analysis of
`the influx and efflux dynamics of the dye and its
`fluorescent intensity.
`
`Materials and Methods
`
`The Test
`A total of seven Sprague-Dawley rats (weigh­
`ing approximately 300 gm) (Central Labora­
`tory for Experimental Animal Studies, Aachen,
`Germany) were anesthetized with a subcutane­
`ous injection of 0.4 ml Hypnorm (Jansen
`GmbH, Neuss, Germany). After complete re­
`moval of abdominal hair, axial-pattern skin
`flaps, modified according to a model by Vidas
`et al.,46 were raised. Measuring 2X8 cm2, they
`were based on the left inferior epigastric neu­
`rovascular bundle (Fig. 1). Before returning the
`flaps to their wound beds, collagen sheets were
`sutured in to delay revascularization from the
`wound bed and its edges. The right femoral vein
`was punctured for injection of indocyanine
`green. An in vitro study demonstrated maxi­
`mum fluorescence at plasma dye concentra­
`tions that indicated a dose of 0.5 mg/kg of body
`weight. The animals were divided into two
`groups. Group I was studied on the day of op­
`eration and on the first and third postoperative
`days. The operation day and days 2 and 4 were
`selected for studying group II. On the seventh
`postoperative day, the viability of the skin flap
`was assessed clinically in all the study animals.
`Room temperature was maintained at 25°G dur­
`ing the entire study period, and standardized
`conditions (focusing of the camera, distance of
`
`plastic AND RECONSTRUCTIVE SURGERY, December 1995
`the object from the camera or the light source,
`etc.) were strictly adhered to.
`
`Indocyanine Green Angiography
`For indocyanine green angiography, a
`2000-W halogen lamp was used to excite the dye
`(Strand Lighting GmbH, Wolfenbuttel, Germa­
`ny) . Because of the considerable development
`of heat, a water filter was included to protect the
`excitation filter 750-FS 40 (LOT, Darmstadt,
`Germany). A long-pass filter RG 850 (LOT, Ger­
`many) was used as a barrier filter. The emission
`of the excited fluorescent dye was then re­
`corded with a Sanyo CCD camera, a videotimer
`(VTG-33, FOR A), a Sony U-Matic videore­
`corder, and a high-resolution monitor. The sig­
`nal was entered into a digital image processing
`system, where it was further evaluated. The test
`construction is depicted in Figure 2.
`
`Digital Image Processing and Evaluation of Indocya­
`nine Green Angiography
`The pictures received from indocyanine
`green angiography were digitalized by a 768 X
`512 pixel, 8-bit image processor (VP 1100-768-
`E-AT; OFG). A series of images was stored at a
`rate of two per second during the first 25 sec­
`onds. Subsequent rates were slower as a result
`of computer capacity. Processing of the digi­
`talized images was then undertaken with a Soft­
`ware Bisc Optimas (Stemmer, Puchheim, Ger­
`many) . Ten different regions of interest were
`defined on the skin flap, and a mean fluores­
`cent intensity was measured over time (see Fig.
`1). The regions of interest were numbered 1 to
`10 from proximal to distal to facilitate evalua­
`tion. The evaluation of indocyanine green an­
`giography was then performed with a model
`that took into account influx and efflux as well
`as lag time, which allowed for the dye to spread
`from the injection site to the region of interest.
`The formula for the time course of the mean
`intensity/(9 is as follows:
`
`where
`fmax = maximum intensity
`f = t - flag
`£]ag = influx lag time
`Cinf = influx time constant
`Ceff = efflux time constant
`To determine the time constants, fmax was set
`at fmzx — max [/(t) ] under the assumption that
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 3 of 14
`
`

`

`Vol 96, No. 7 / MICROCIRCULATION OF AN AXIAL SKIN FLAP
`
`1639
`
`collagen sheet
`
`l inf* epigastric
`neurovasc* pedicle
`
`Fig. 1. A 300-gm Sprague-Dawley rat with a skin flap (2 X 8
`Beneath the flap is a synthetic collagenous sheet.
`
`cm2) based on the left inferior epigastric artery, vein, and nerve.
`
`Ccff is considerably larger than Cinf, that is, Ceff §>
`Cinf. Calculation of the influx constant Cinf re­
`sults from
`* < 4nax
`g(
`After finding the logarithms of the data up to
`the peak of the curve at t =■ £max, a straight line
`is drawn by linear regression. The slope of the
`straight line provides the influx time constant.
`The efflux time constant is calculated in a sim­
`ilar fashion. In this case, however, fm3X is not
`
`subtracted because these data follow the peak.
`The time constants thus calculated (Cinf, Ceff)
`and the maximum intensity
`then allow
`conclusions to be drawn on the state of blood
`circulation. The quality of the calculated data
`was confirmed by determining the correlation
`coefficients.
`The Dye: Indocyanine Green (Cardio Green)
`Indocyanine green is supplied in 25- and
`50-mg single packs (Paesel Company, Frank-
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 4 of 14
`
`

`

`1640
`
`PLASTIC AND RECONSTRUCTIVE SURGERY, December 1995
`
`Fig. 2. Indocyanine green angiography. The dye was excited after its intravenous application
`and distribution in the bloodstream. The emitted light is selected out by the barrier filter and
`recorded by the CCD camera. The resulting images are stored and processed (1, 2000-W halogen
`lamp; 2, water filter; 3, barrier filter; 4, water pump; 5, excitation filter; 6, CCD camera; 7,
`videorecorder, videotimer, monitor; 8, digital image processing system).
`
`furt, Germany). As a finished solution, indocya­
`nine green contains 5.0 to 9.5 percent sodium
`iodine. For this reason, the manufacturer rec­
`ommends caution when applying it in patients
`with iodine allergy or thyroid disease. Similarly,
`administration during pregnancy or lactation is
`not guaranteed free of risk. The incidence of
`untoward reactions to indocyanine green is low.
`The literature cites a rate of 1 in 42,000.46 An
`article by Carski et al.4V mentions a case number
`of 4 reactions in over 240,000 applications.
`These reactions usually take a mild and harm­
`less course with symptoms such as nausea, hot
`flushes, headache, and urticaria. On the other
`hand, exceptional cases also have been de­
`scribed with more severe side effects such as
`dyspnea, edema, fall in blood pressure, and
`tachycardia.46-48 Michie et al.49 report an in­
`creased incidence of reactions to indocyanine
`green in patients with chronic uremia. On the
`whole, side effects from indocyanine green are
`not considered genuinely allergic but rather a
`pseudoallergic reaction.50,51 The lack of eosin-
`ophilia and no significant IgE increase support
`this view.50 When using the fluorescent dye in­
`docyanine green in clinical practice, however,
`the possibility of a reaction should be kept low
`or even excluded, so preventative measures or
`
`a reliable test to initiate prophylaxis deserves
`consideration. As a rule, atopic individuals and
`patients belonging to these groups (those with
`hay fever, allergic eczema, etc.) are considered
`at risk. A positive intracutaneous test provides
`reliable results, while a negative result would
`not exclude a later reaction. The intravenous
`injection of a small test dose would be the safest
`method of excluding sensitization, although it
`could interfere with the subsequent examina­
`tion.
`Before the examination, an antihistamine or
`a cortisone preparation could be applied as a
`direct form of prophylaxis. An important pre­
`requisite for using this dye for fluorescence is an
`exact knowledge of its spectral properties. In
`serum, indocyanine green displays an absorp­
`tion maximum at 805 nm and an emission max­
`imum at 835 nm.
`
`Results
`After intravenous injection of the dye, a rapid
`inflow was observed by means of the afferent
`vessel of the axial-pattern skin flap. The first
`fluorescence was registered after 1.20 s (SD =
`0.54 s). Comparing the individual test days, vas­
`cularity was sparse on the operation day, while
`the following postoperative days showed clear
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 5 of 14
`
`

`

`Vol 96, No. 7 / MICROCIRCULATION OF AN AXIAL SKIN FLAP
`
`1641
`
`Fig. 3. Indocyanine green angiography. {Above, left) Second postoperative day, the axial skin flap after 1.45 s. {Above, right)
`Second postoperative day, the axial skin flap after 1.98 s. {Below, left) Second postoperative day, the axial skin flap after 4.06 s.
`{Below, right) Second postoperative day, the axial skin flap after 1.00 min.
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 6 of 14
`
`

`

`1642
`
`PLASTIC AND RECONSTRUCTIVE SURGERY, December 1995
`
`1 l f 1 a \
`
`ROI 1 ROI2 ROI 3 ROI4 ROI5 ROI 6 ROI7 ROI 8 ROI 9 ROI10
`
`■ Operation day ■■ l.post OP E32. post OP (883.post OP
`mi 4. post OP
`
`SD (standard deviation)
`
`Operation
`day
`1,1
`2,2
`1,8
`1,7
`1,2
`1
`1,5
`2,5
`2,1
`1,6
`
`ROI 1
`ROI 2
`ROI 3
`ROI 4
`ROI 5
`ROI 6
`ROI 7
`ROI 8
`ROI 9
`ROI 10
`
`l.post OP 2. post OP 3.post OP
`
`4.post OP
`
`0,9
`0,2
`0,3
`0,1
`0,1
`0,2
`0,2
`0,6
`0,5
`0,9
`
`0,2
`0,2
`0,1
`0,1
`0,1
`0,3
`0,3
`1,1
`0,7
`0,6
`
`0,4
`0,4
`0,6
`0,6
`0,8
`0,8
`1
`0,9
`1
`2,3
`
`0,5
`0,4
`0,7
`0,7
`0,8
`0,8
`1,1
`1,2
`1,1
`1,6
`
`Fig. 4. Influx dynamics in the axial skin flap during the postoperative course.
`
`vascular structures during the influx phase (Fig.
`3, above, left and right). In the postoperative
`course, the size of the structures visible in the
`skin flap also increased. Depending on the size
`of the vascular network, the proximal part of the
`flap very rapidly became homogeneously fluo­
`rescent (see Fig. 3, below, left). The distal partwas
`subject to a slower distribution of the dye. After
`the flap was completely stained (see Fig. 3, below,
`right), additional vascular structures appeared
`during the efflux phase that differed from those
`seen during the influx phase. This vascularity
`persisted for a much longer time.
`Evaluations of indocyanine green angiogra­
`
`phy for the operation days showed a very con­
`spicuous microcirculatory state. On the whole,
`a retarded influx was registered that manifested
`itself particularly in the proximal and distal
`parts of the flap (Fig. 4j. With varying degrees
`of expression, the efflux pattern was visible over
`longer periods of time across the entire flap.
`The maximum fluorescence intensities showed
`low values. The development in the postoper­
`ative course produced a homogenization of the
`flow dynamics, allowing an increasing improve­
`ment in influx and, in particular, in efflux dy­
`namics in indocyanine green angiography. On
`the first postoperative day, a very clear rise in
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 7 of 14
`
`

`

`Vol 96, No. 7 / MICROCIRCULATION OF AN AXLAL SKIN FLAP
`
`1643
`
`ROI1 ROI2 ROI 3 ROI 4 ROI5 ROI 6 ROI7 ROI 8 ROI 9 ROI10
`■Operation day Ml.post OP E32. post OP E33.post OP
`H4. post OP
`
`SD (standard deviation)
`
`l.post OP 2. post OP 3.post OP
`
`4.post OP
`
`17,8
`14,9
`37,7
`45,3
`127
`124,8
`566,8
`975,6
`995,3
`30
`
`78,8
`71
`55,2
`65
`86
`43,4
`22,6
`279,5
`56,9
`600,3
`
`30,6
`31,2
`40
`37,9
`33,8
`33,3
`30,6
`23,7
`45,5
`169,2
`
`25,4
`39,9
`47,2
`46,6
`93,7
`155,6
`118,4
`136,3
`144,5
`115,2
`
`Operation
`day
`936,7
`147,2
`787,8
`345
`995,4
`1471,3
`466,9
`3078,9
`147,7
`180,7
`
`ROI I
`ROI 2
`ROI 3
`ROI 4
`ROI 5
`ROI 6
`ROI 7
`ROI 8
`ROI 9
`ROI 10
`
`Fig. 5. Efflux dynamics in the axial skin flap during the postoperative course.
`
`the efflux constants was shown from the prox­
`imal to the distal part of the flap (Fig. 5). This
`tendency, which increased toward the tip of the
`flap, declined over the following days and was
`only slightly pronounced by day 4. In the same
`way, the efflux constant took on a course similar
`to the rising tendency of the influx (see Fig. 4).
`The reverse was true for the behavior of the
`maximum fluorescence intensity. The fluores­
`cence values that declined toward the flap tip
`on the first postoperative day rose over time,
`displaying only a slight gradient by day 4 (Fig.
`6). A final examination at the end of the trial
`
`revealed clinically viable skin flaps in all ani­
`mals.
`
`Discussion
`When assessing the microcirculation of skin
`or a skin flap, its complexity demands the use of
`a technique that will provide as objective and
`reliable a result as possible. The technique of
`the fluorescein test described by Lange and
`Boyd12 is fundamentally simple to use for ex­
`amining the microcirculation of tissue. The ex­
`cellent studies by these authors using the dye
`sodium fluorescein showed a good correlation
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 8 of 14
`
`

`

`1644
`
`plastic AND reconstructive surgery, December 1995
`
`200
`
`150
`
`100
`
`50
`
`0
`
`ROI1 ROI2 ROI 3 ROI4 ROI5 ROI 6 ROI7 ROI 8 ROI9 ROI10
`
`Operation day HI.post OP E32. post OP E33.post OP
`!4.post OP
`
`SD (standard deviation)
`
`Operation
`day
`15,6
`19,1
`12,1
`16,6
`14,7
`18,4
`25
`23,4
`17,7
`14,3
`
`ROI 1
`ROI 2
`ROI 3
`ROI 4
`ROI 5
`ROI 6
`ROI 7
`ROI 8
`ROI 9
`ROI 10
`
`l.post OP 2. post OP 3.post OP
`
`4.post OP
`
`5,9
`5,4
`5,4
`5,3
`5,9
`7
`7,7
`17,7
`10,1
`4,6
`
`3,2
`8,4
`4,1
`6,9
`5
`2,9
`2,3
`2,1
`7,2
`10,1
`
`6
`8,3
`7,4
`6,7
`8,7
`10,9
`6,9
`8,9
`7,6
`2,9
`
`2,5
`3,6
`6,3
`9,2
`5,3
`5,4
`4,7
`3,9
`4,5
`2,9
`
`FlG. 6. Maximum fluorescence intensity in the axial skin flap during the postoperative course.
`with the circulation in skin flaps. The pharma­
`remaining intravasal, allowing conclusions to
`cokinetic properties of sodium fluorescein,
`be drawn on the existence of a perfused vessel.
`however, render it only partially suitable for
`Furthermore, the short half-life of indocya­
`nine green is enormously advantageous for its
`clinical use. The possibility of its application is
`clinical application. According to Meijer et al.,42
`particularly limited by the diffusion of the dye
`indocyanine green is subject to biphasic elimi­
`into the interstitium and its remaining in the
`nation from plasma. Over 90 percent of the
`perivascular tissue for some time. After the in­
`applied dose of 0.5 mg/kg of body weight is
`troduction of indocyanine green, however, a
`vital dye has made its entry into the fluores­
`washed out during the initial phase with a half-
`life of 3 to 4 minutes. This allows short intervals
`cence technique that promises an improved
`between each examination for assessing the mi­
`clinical usage on the grounds of its physiologic
`crocirculation of the tissue. Changes in circu­
`properties. A primary aspect of its excellent be­
`lation occurring acutely, such as arterial occlu­
`havior is its almost complete binding to plasma
`sion, vascular thrombosis, or torsion of the
`proteins. This guarantees the fluorescent dye
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 9 of 14
`
`

`

`Vol. 96, No. 7 / MICROCIRCULATION OF AN AXIAL SKIN FLAP
`Greyscale
`
`1645
`
`A ROI 1 a ROI 2 o ROI 3 a ROI 4
`
`& ROI 5
`
`© ROI 6
`
`® ROI 7 o ROI 8
`
`a ROI 9
`
`■ ROI 10
`
`T (sec)
`
`Fig. 7. Original and calculated indocyanine green angiographic data of the axial skin flap depicted in Figure 3 on the second
`postoperative day. Original data stippled, corresponding calculated data from 4 values per second lined.
`
`pedicle, can therefore be detected shortly after
`a previous negative examination. With sodium
`fluorescein, the dye used until now, short-term
`controls would be useless because of the per­
`sistence of the dye in the tissue. A fluorescent
`area might in actual fact be less perfused or even
`not perfused at all.
`Indocyanine green offers another advantage
`for assessing the dynamics of microcirculation.
`The analysis of dye distribution over time with
`a dye that remains exclusively intravasal will al­
`low reliable conclusions to be drawn on the
`dynamics of blood flow. With sodium fluores­
`cein, the additional dimension of transcapillary
`diffusion complicates any such conclusions and
`makes assessment more difficult.
`The spectral properties of indocyanine green
`are also considered a positive factor. The ab­
`sorption and emission maxima of indocyanine
`green are to be found in the near-infrared area.
`It is in just this range that human skin has its
`“optic window.” The literature cites a penetra­
`
`tion depth of approximately 3 mm for this wave­
`length.52
`The indocyanine green angiography exami­
`nations in the present study demonstrated per­
`fusion of the skin flap by means of the afferent
`inferior epigastric artery immediately after op­
`eration. The time lag before the first appear­
`ance of fluorescence gives an impression of the
`macro circulatory state of the organism. Here
`initial fluorescence was demonstrated after 1.20
`s (SD = 0.54 s)—a value that may well be in­
`terpreted as a good macro circulatory state of
`the study animals. On further evaluation of the
`postoperative days, a state of hemodynamic im­
`balance was noted in the microcirculatory sys­
`tem of the flap. The disturbed microcirculation
`might well be due mainly to the loss of sympa­
`thetic innervation following flap dissection.1
`Adrenergic afferences control vessel diameter
`and the effective flow pressure gradient. Both
`factors, together with the viscosity of the blood,
`represent the most important parameters in
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVADAQ TECHNOLOGIES
` Page 10 of 14
`
`

`

`1646
`capillary circulation. Sympathectomy subse­
`quently causes dilatation of the vessels and a
`reduction in the perfusion pressure gradient.1
`However, further interpretation of the postop­
`erative days would seem, at the moment at least,
`very difficult. Nevertheless, the influx and efflux
`constants, as well as the maximum intensity,
`show values that are quite plausible from a the­
`oretical physiologic point of view. It is essential,
`however, that the efflux constant is not con­
`fused with the outflow rate by means of the
`efferent vascular system. The efflux constant
`contains information that includes the recircu­
`lation of the dye, possible adhesion to the vessel
`wall, and the small proportion of transcapillary
`penetration. For this reason, Ceff is many times
`larger than Cinf. To a certain degree, the influx
`constant is related to the flow rate of the affer­
`ent vascular system.
`As McCarthy states,1 the skin at this point of
`time can no longer be regarded as a thermo­
`regulatory organ but rather as tissue that is just
`trying to survive by humoral, metabolic, and
`physical mechanisms. If the organism does not
`succeed in establishing a new hemodynamic
`equilibrium, then within a certain period of
`time the tissue will die. In the further postop­
`erative course, the effects of this compensation
`become apparent, effects that lead to changes
`in perfusion and were then detected by indo­
`cyanine green angiography. Adaptation of the
`hemodynamics to the new situation created a
`microcirculatory state that became more homo­
`geneous during the first postoperative day. Ini­
`tially, however, the distal part of the flap still
`demonstrated reduced perfusion. This is re­
`flected in the increase in the influx and efflux
`constants and the fall of the maximum intensity
`values. Situated furthest from the base of the
`flap, this region is usually one of the most crit­
`ical areas as far as blood supply is concerned.
`From day 3 to 4 onward, anatomic changes
`then exert their influence on the microcircu­
`lation. Long years of experience support this
`fact, which again was recognized in the present
`study by the increasing vascularity during the
`influx phase. Blood circulation is improved, on
`the one hand, by the growth of new capillaries
`from the surrounding tissue and, on the other
`hand, by the dilatation of already available anas­
`tomoses between vessels running in a transverse
`direction. An increase in the actual density of
`the vessels does not occur.53 Revascularization
`from surrounding tissue was prevented in our
`model by a synthetic collagenous sheet. The
`
`PLASTIC AND RECONSTRUCTIVE SURGERY, December 1995
`further improvement in flap circulation up to
`day 4 ought therefore to be due to increased
`perfusion through already present anastomo­
`ses. The fact that another type of vascular struc­
`ture becomes visible during the efflux phase is
`possibly due to the appearance of the venous
`system. This conclusion is reached by the fact
`that the structure of the vessels differs from that
`of the arterial influx phase. Why the venous
`system is depicted for a much longer time than
`the arterial system is still unexplained. Binding
`of indocyanine green to glycocalix or directly to
`endothelial cells is one of the possible expla­
`nations offered by Bollinger et al.3 with respect
`to fluorescence videomicroscopy. Phagocytosis
`of the protein-bound dye in endothelial cells
`has not yet been excluded. The influence of the
`non-protein-bound component of indocyanine
`green certainly deserves consideration. The
`literature cites a proportion of 3 to 5 per­
`cent 3°-33>38-4L54
`Prognoses based on the present study regard­
`ing viability should only be made with caution.
`Marginal or normative values cannot be deter­
`mined because of the small number of cases.
`The time interval between each examination
`was probably too long. Determining the optimal
`point of time for examination must be the ob­
`ject of a further study; the ischemic time of the
`tissue necessitates shorter intervals during the
`first postoperative hours. A larger animal study
`or clinical trials should be considered here.
`Conclusions
`Indocyanine green angiography is an excellent
`method for recording the microcirculatory state
`of a given tissue. Computerized analysis of the
`distribution pattern of the tracer, which is almost
`completely bound intravasally, offers the best pos­
`sible correlation with the actual pattern of perfu­
`sion. The distribution of perfusion in the dimen­
`sion of space is also guaranteed by the macro­
`scopic recording technique. Guaranteeing the di­
`mension of time is certainly more difficult. In
`order to capture the sensitive regulation of the
`microcirculatory system, the procedure should of­
`fer the possibility of continuous analysis over time,
`something that indocyanine green angiography is
`most probably unable to fulfill. A quick repetition
`of the examination, however, is guaranteed by the
`short half-life of indocyanine green. The first of a
`number of further studies suggests a minimal time
`interval between examinations of approximately
`30 to 45 minutes. If the microcirculation is grossly
`disturbed, however, the intervals can be length­
`
` VISIONSENSE - 1008
` VISIONSENSE v.
` NOVA