Reperfusion of Free Full-Thickness Skin Grafts in Periocular Reconstructive Surgery Monitored Using Laser Speckle Contrast Imaging

Purpose:

Free skin grafts are frequently used in reconstructive surgery. However, little is known about the course of reperfusion due to the previous lack of reliable perfusion monitoring techniques. The aim of this study was to use state-of-the-art laser speckle contrast imaging to monitor free skin grafts in the periocular area.

Methods:

Seven patients needing surgery due to tumor removal or cicatricial ectropion in the periocular region underwent reconstructive surgery using free skin grafts from either the contralateral upper eyelid or the upper inner arm. The free skin grafts measured 10–30 mm horizontally and 9–30 mm vertically. Blood perfusion was monitored using laser speckle contrast imaging immediately postoperatively (0 weeks) and at follow-up after 1, 3, and 7 weeks.

Results:

All grafts were reperfused gradually during healing, the median value being 46% in the central part of the graft after 1 week and 79% after 3 weeks. The grafts were completely reperfused after 7 weeks. No difference was observed in the rate of reperfusion between the center and periphery of the grafts (p = not significant). The cosmetic and functional outcome was excellent in all but 1 patient, who developed ectropion that had to be surgically corrected.

Conclusions:

Skin grafts in the periorbital area are fully reperfused after 7 weeks. The periocular area is known to be well-vascularized and thus forgiving to reconstructive surgery. Future investigations of the reperfusion of free skin grafts in other parts of the body or in higher-risk populations should be carried out.

Free skin grafts in the periocular area are fully reperfused after 7 weeks. The periocular area is known to be well-vascularized.

The tissue available in reconstructive surgery in the periocular area is often limited, and it may be difficult to ensure wound coverage. In cases where primary closure or closure using flaps is not possible, free full-thickness skin grafts are often considered. In reconstructive surgery in the periocular area, skin grafts from the inside of the arm, the pre- and postauricular area, the supraclavicular fossa, or the contralateral upper eyelid above the skin crease, are often preferred.¹

The process of reperfusion has been studied since the latter half of the 1800s.^2–4 However, the process of reperfusion of skin grafts is still not fully understood. Perfusion monitoring has long been performed through clinical examination, as first described by the Italian Renaissance surgeon, Gaspare Tagliacozzi, that is, by feeling the temperature of the skin, observing the color, and measuring the capillary refill time.⁵ This kind of clinical examination is rapidly and easily performed, and is still used by surgeons worldwide. However, the method is highly subjective and dependent on the experience of the surgeon. Technological developments have led to other observer-independent methods of evaluating perfusion. One of the most widely used is fluorescence angiography, a technique first introduced by Lange and Boyd in 1943⁶ to study blood perfusion in peripheral vascular diseases. Sodium fluorescein was originally used as the fluorescent agent, but in 1973 Flower and Hochheimer introduced indocyanine green into the fluorescence technique.⁷ The patient is given an intravenous injection of indocyanine green, and an image of the perfusion is obtained using an infrared camera. However, this technique is invasive, and is not suitable for monitoring perfusion over time. Thermal imaging is a noninvasive technique in which an infrared camera is used to detect heat. However, there is a delay between the change in perfusion and the change in temperature, and other mechanisms apart from perfusion may lead to changes in the temperature of the tissue, for example, other metabolic processes in the cells such as inflammatory responses. Methods such as intravenous injections of radioisotopes have been used in animal models,⁸ and the examination of histological sections after skin grafting has been performed in humans,⁹ however, these methods are not suitable in clinical practice.

Laser-based techniques have recently emerged, enabling the noninvasive monitoring of perfusion in the clinical setting. In laser speckle contrast imaging (LSCI) the skin area of interest is illuminated by infrared (785 nm) laser light. Dark and bright areas are created by interference of the light backscattered from moving particles in the illuminated area, creating a speckled pattern. This pattern is recorded in real-time by a camera, and the perfusion is automatically calculated by the system by analyzing the variations in the speckle pattern created by moving particles. It is generally believed that the measurement depth of LSCI is 300 μm. After transplantation changes take place in both the wound bed and the skin graft. According to microscopic observations in mice, preexisting graft vessels are believed to act as nonviable conduits during the revascularization process.¹⁰ One disadvantage of the LSCI technique is that the signal does not discriminate between the formation of capillaries and a change in the microvascular architecture of the graft and changes in the dimensions of capillaries leading to a change in blood flow. Neither do we know how the change in the papillary networks in the dermal-epidermal junction of skin after skin transplantation affect the measurement. We recently used LSCI to monitor the reperfusion of free, full-thickness skin grafts used as the anterior lamella in the modified Hughes procedure, where the free graft is placed on a tarsoconjunctival flap.¹¹ However, most free skin grafts in clinical practice are placed on a vascularized wound bed, rather than on a flap,¹² and the reperfusion of these kinds of grafts has not previously been studied using modern imaging technique. This study was performed to monitor the reperfusion of free skin grafts when placed on a vascularized wound bed in the periocular area, using LSCI.

METHODS

Ethics.

Ethical approval was obtained from the Ethics Committee at Lund University, Sweden. The research followed the Declaration of Helsinki as amended in 2008. Fully informed consent was obtained from all patients included in the study.

Subjects.

A total of 7 patients, 2 women and 5 men, were included between October 2016 and March 2020. Three of the patients required reconstructive surgery following the excision of a basal cell carcinoma, and 2 following the excision of a squamous-cell skin carcinoma. Two patients had developed cicatricial ectropion; one caused by previous trauma and the other by previous tumor excision with successive shortening of the anterior lamella. The median age was 86 years (range 60–92 years). One patient was taking antihypertensive medication and 1 patient suffered from diabetes. Two patients with cardiovascular conditions were taking anticoagulants. None of the participants was currently a smoker, but 4 patients were former smokers who had quit smoking more than 5 years ago. No patient had previous radiotherapy in the periocular area. More skin grafts were performed during this period, but due to logistic reasons, the authors were not able to include them all.

Surgical Procedure.

Surgery was performed under local anesthesia using lidocaine (20 mg/ml) (Xylocaine; AstraZeneca, Södertälje, Sweden). Epinephrine was avoided because of its vasoconstrictive effect, which would have influenced the LSCI measurements. The full-thickness skin graft was harvested from the contralateral upper eyelid (n = 4) or the inside of the upper arm (n = 3). After harvesting the graft, underlying adipose tissue and hair follicles were removed. The graft was then trimmed so as to be 1.5 times larger than the recipient site, to avoid wound contraction. The free skin grafts measured 10–30 mm horizontally and 9–30 mm vertically. The grafts were used to cover defects on the lower eyelid (n = 3), the upper eyelid (n = 2), the median canthal area (n = 1), or the upper cheek (n = 1). The graft was secured with continuous and/or simple interrupted 6-0 non-resorbable (Ethilon 6-0; Ethicon, Somerville, NJ) or resorbable 6-0 sutures (Vicryl 6-0; Ethicon). Great care was taken to minimize the tension when suturing the graft in place. Stab incisions were made in grafts larger than 15 mm to allow for drainage of fluid and adhesion to the recipient bed. A secure pressure dressing was applied over the surgical site to ensure that the graft maintained direct contact with the underlying wound bed, and to immobilize it during healing. The cotton bolster, secured with Tegaderm (3M Company, St. Paul, MN), remained in place until days 2–3. Great care was taken to avoid pressure on the graft, as it would most likely influence reperfusion. Frost sutures, using 4-0 non-resorbable sutures (Silk 4-0, Ethicon) were used in all patients receiving free skin grafts in the lower eyelid to provide upward tension, and were also left in place for 2–3 days. The patients were told to avoid trauma to the site and strenuous activity for 2 weeks after surgery. The sutures were removed after 1 week. Surgery was performed by 2 experienced senior surgeons at the Department of Ophthalmology at Skåne University Hospital, Lund, Sweden.

Perfusion Measurements.

Blood perfusion was monitored using a PeriCam PSI NR System (Perimed AB, Stockholm, Sweden). Perfusion was monitored immediately postoperatively (denoted 0 weeks) and on 3 follow-up occasions. Due to logistic reasons, the time of the follow-up visits varied. The data were therefore grouped into the following time intervals: follow-up at 7–8 days (denoted 1 week), 21–28 days (denoted 3 weeks), and 39–54 days (denoted 7 weeks).

Calculations and Statistics.

When using LSCI, perfusion is expressed in arbitrary units, that is, perfusion units. The results were normalized to the value of perfusion obtained at a reference point in intact skin just outside the graft (which was set to 100%) (Figure 1). The reference point was infiltrated with local anesthesia in the same way as the surgical area. The perfusion at the reference point was measured at each follow-up visit to the clinic. Zero perfusion values were obtained from the graft immediately after suturing it in place. At this stage, the graft is completely avascular, and the perfusion values obtained are the result of a combination of movement artifacts and the biological zero, that is, the signal obtained from tissue in the absence of vascular flow. These immediate postoperative values were therefore set to 0%, and were subtracted from the perfusion values obtained at postoperative follow-up.

FIG. 1.

Photograph of a free full-thickness skin graft immediately after surgery (A). Regions of interest (ROI) used for perfusion measurements are shown at the center (white circle) and along the periphery (white rectangle) of the graft. Perfusion was expressed as the percent of perfusion at a reference point just outside the graft (black circle). (B) The same graft 1 week after surgery.

GraphPad Prism 7.0a (GraphPad Software Inc., San Diego, CA) was used for calculations and statistical analysis. Results are expressed as median values with 95% confidence intervals. Statistical analysis was performed using the Kruskal-Wallis test with Dunn’s test for multiple comparisons. Significance was defined as: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and p > 0.05 (not significant).

RESULTS

The perfusion results obtained with the PeriCam showed that the free skin grafts were rapidly reperfused (Figure 2). After 1 week the perfusion was 46% in the central part of the graft, and after 3 weeks 79%. However, after 3 weeks, 2 of the grafts were fully reperfused. Complete reperfusion, that is, that did not differ significantly from that in the surrounding tissue, was seen after 7 weeks (p = not significant). No difference was observed between the perfusion at the center and along the periphery of the graft, indicating revascularization of the entire graft from the underlying wound bed.

FIG. 2.

Above: a scatterplot showing the blood perfusion in the free skin grafts, immediately postoperatively (0 weeks), and at follow-up after 1, 3, and 7 weeks. Perfusion was measured at the center and along periphery of the grafts. Data are expressed as the percentage (median values and range) after normalization to the perfusion at a reference point just outside the grafts (100%). It can be seen that reperfusion occurred simultaneously in the center and periphery of the graft, and that complete reperfusion was achieved after 7 weeks. Below: representative examples of laser speckle contrast images.

All free skin grafts survived and healed well. There was no sign of ischemia or pressure necrosis. Both the cosmetic and functional outcome was excellent in all but 1 patient, who developed postoperative cicatricial ectropion of the lower eyelid, which was later surgically corrected. No complications were reported at the donor sites.

DISCUSSION

To the best of the authors’ knowledge, this is the first study of the reperfusion of free skin grafts on a vascularized bed in the periocular region. We found that revascularization took place rapidly; reaching 79% in 3 weeks in the central part of the graft, and complete reperfusion after 7 weeks. Reperfusion of free full-thickness grafts has previously been studied in animals and other regions of the human body, using other techniques than LSCI. The process of revascularization takes place through anastomoses between blood vessels of the graft and the recipient, known as inosculation.^13,14 Angiogenesis also takes place in the wound bed, and the new blood vessels invade the graft (ingrowth). Preexisting graft vessels have been shown to act as nonviable conduits for the invasion of new vessels.¹⁰

Skin graft reperfusion has mainly been studied using microscopy and histological examinations. In 1956, Converse and Rapaport¹⁵ used microscopy to study the results of full-thickness grafting on the radial aspect of the volar surface of the forearm in humans and found sluggish flow in the vessels in the graft on the third day postoperatively. Cyanotic discoloration was observed during the first 6–7 days, suggesting poor blood flow. These observations are consistent with our finding that the free full-thickness grafts are reperfused rapidly. Rapid reperfusion was also reported by Ohmori and Kurata⁸ in 1960 using intravenous injections of radioisotopes in a rabbit model, showing blood flow and isotope uptake in the graft 4 days postoperatively. In 1968, Clemmesen and Ronhovde⁹ took biopsies from humans after skin grafting, and found dilated vessels connected to the original vessels of the graft 3 days later. More recently, in 2006, Capla et al.¹⁶ observed vascular ingrowth in the periphery of full-thickness grafts in a mouse model on day 3 through histological examination. In 2008, Lindenblatt et al. described a mouse model that allowed continuous monitoring of the microcirculation during skin graft healing using repetitive intravital microscopy. They reported that capillary buds and sprouts were visible on day 2, and that the graft capillaries contained blood on day 3. They concluded that the original skin microcirculation was almost completely restored on day 5.¹⁷ Animal models such as the ones described above have been used in many studies on the healing of full-thickness skin autografts, but differences between the vascular systems in animals and humans prevent direct comparisons. This might explain the difference in time between those studies and the present one. The periocular area is known to be forgiving in reconstructive surgery, which may be due to the rich vascularization of this region. It should, however, be borne in mind that the skin grafts in this study were very thin, that is, only 2–3 mm thick. Skin grafts in other regions of the body may be thicker, which might influence the rate of reperfusion and the survival of the graft.

The LSCI signal was somewhat reduced in the weeks after surgery. It cannot be deduced from the present study whether this reduction was due to reduced perfusion in the graft, or because the graft was thicker than normal, non-transplanted skin. The graft is generally designed to be larger than the defect, and hence somewhat folded in order to allow shrinkage in the postoperative phase.

Several factors are thought to influence the survival of grafts; the underlying graft bed being one of the most important. Grafts placed on well-vascularized tissue are believed to be more likely to survive than those on a less vascularized tissue. A graft placed on a non-vascularized tissue, such as bone without periosteum, is not expected to survive.¹⁸ Grafts placed on poorly vascularized wound beds have been found to be ischemic for longer periods than grafts placed on wound beds with better blood perfusion.¹⁹

When studying skin grafts in the modified Hughes procedure 2019, we found that the grafts overlying the tarsus was reperfused at the same rate, and survived as well as the grafts applied to orbicularis muscle.¹¹ We attributed these observations to the rich vascularization of the periocular area, and the fact that the grafts had been soaked in tear fluid, which is known to have the similar spectrum of nutrients as the blood.²⁰

The results of the study are mainly of interest to ophthalmic plastic surgeons. The main limitation of the present study is the limited number of participants, which did not allow subgroup analyses. Therefore, factors known to affect blood perfusion and compromise the oxygenation of tissue, such as smoking, diabetes, previous surgery, or radiation therapy, could not be taken into account. Furthermore, the size of the graft varied considerably between patients, which could have influenced the outcome. Again, the small number of patients did not allow any analysis of the effects of graft size on reperfusion. Another limitation of the present study is that perfusion measurements could not be made on a daily basis since it was impractical. The study was conducted in an outpatient clinic, and many of the patients had to travel far to attend follow-up.

In conclusion, free full-thickness skin grafts in the periocular area were found to be completely reperfused after 7 weeks. The periocular area is known to be well-vascularized and thus forgiving to reconstructive surgery. Future investigations should focus on the reperfusion of free skin grafts of different sizes, in other parts of the body, or in higher-risk populations.

ACKNOWLEDGMENT(S)

We would like to thank all the surgical staff involved at the Department of Ophthalmology, Skåne University Hospital, for their kind cooperation in this work. Helpful advice on language and linguistics was given by Helen Sheppard.